_____________________________________________________________________________________________________________

Fam Proc 13:141-168, 1974

Brief Therapy: Focused Problem Resolution

JOHN H. WEAKLANDa
RICHARD FISCH, M.D.a
PAUL WATZLAWICK, PH.D.a
ARTHUR M. BODIN, PH.D.a
aBrief Therapy Center, Mental Research Institute, Palo Alto, California.

This article describes a general view of the nature of human problems and their effective resolution and of related
specific procedures, growing out of our prior work in family therapy, that have developed during six years of research
on rapid problem resolution. With treatment limited to a maximum of ten sessions, we have achieved significant success
in about three-fourths of a sample of 97 widely varied cases, and this approach to problems appears to have
considerable potential for further development and wider application.
In the last few years, brief treatment has been proliferatingboth growing and dividing. As Barten's (2) recent collection
of papers illustrates, "brief therapy" means many different things to many different therapists. The brief therapy we wish to
present here is an outgrowth of our earlier work in that it is based on two ideas central to family therapy: (a) focusing on
observable behavioral interaction in the present and (b) deliberate intervention to alter the going system. In pursuing these
themes further, however, we have arrived at a particular conceptualization of the nature of human problems and their
effective resolution, and of related procedures, that is different from much current family therapy.
We have been developing and testing this approach at the Brief Therapy Center over the past six years. During this
period the Center, operating one day a week, has treated 97 cases, in which 236 individuals were seen. (We have also had
extensive experience using the same approach with private patients, but these cases have not been systematically followed
up and evaluated.) These 97 cases reached us through a considerable variety of referral sources, and no deliberate selection
was exercised. As a result, although probably a majority of our cases involve rather common marital and family problems,
the sample covers a wide range overall. We have dealt with white, black, and oriental patients from 5 to over 60 years old,
from welfare recipients to the very wealthy, and with a variety of both acute and chronic problems. These included school
and work difficulties; identity crises; marital, family, and sexual problems; delinquency, alcohol, and eating problems;
anxiety, depression, and schizophrenia. Regardless of the nature or severity of the problem, each case has been limited to a
maximum of ten one-hour sessions, usually at weekly intervals. Under these circumstances, our treatment has been
successfulin terms of achieving limited but significant goals related to the patients' main complaintsin about
three-fourths of these cases. We have also demonstrated and taught our approach to a number of other therapists in our
area.
We present our approach here for wider consideration. Any form of treatment, however, is difficult to convey adequately
by a purely verbal account, without demonstration and direct observation. We will, therefore, begin by discussing the
significance and nature of our basic premises in comparison with other forms of treatment. Hopefully, this will provide an
orienting context for the subsequent descriptionsupplemented with illustrative case materialof our interrelated
concepts, plan of treatment, specific techniques, and results.

PSYCHOTHERAPYPREMISES AND PRACTICES

In characterizing treatment approaches, although some over-simplification may result, outlining basic premises may
make their natureand especially, their implicationsmore plain. Often, attention is concentrated on what is explicit and
detailed, while what is common and general is neglected. Yet, the more general an idea, the more determinative of behavior
it isespecially if its existence is not explicitly recognized. This holds for interpersonal influence as well as individual
thinking and behavior; Robert Rosenthal's (21) experiments demonstrate how the beliefs, assumptions, expectations, and
biases of an experimenter or interviewer have a profound effect on his subjects. Similarly, the beliefs and theories held by a
therapist may strongly influence not only his technique but also the length and outcome of his treatmentsby affecting his
patient's behavior, his evaluation of that behavior, or both.
For instance, if schizophrenia is conceptualized as a gradual, irreversible mental deterioration involving loss of contact
with reality, then attempts at psychotherapeutic contact make little sense, and the only reasonable course of action is
long-term hospitalization. The hospitalized patient is then likely to react in a way that clearly justifies this initial
"preventive" action. Alternatively, if schizophrenia is seen as a manifestation of a dysfunctional structure of family
relationships, the outlook is different and more hopeful, although basic restructuring of the family system is now likely to be

1
_____________________________________________________________________________________________________________

seen as necessary. Again, in terms of the postulates of classical psychoanalytic theory, symptom removal must perforce lead
to symptom displacement and exacerbation of the patient's condition, since it deals only with manifestations of deeper
problems. The premises of the theory permit no other conclusion, except the alternative of claiming that the problem must
not have been a "real" one (22). On the other hand, in therapies based on learning or deconditioning theories, symptom
manipulation is consistent with the theoretical premises. This enables the therapist to try very different interventionsand,
to some extent, constrains him to do so.
That is, all theories of psychotherapy (including our own) have limitations, of practice as well as conception, that are
logically inherent in their own nature. Equally important, these limitations are often attributed to human nature, rather than
to the nature of the theory. It is all too easy to overlook this and become enmeshed in unrecognized, circular explanations.
Stating the basic premises of any psychotherapeutic theory as clearly and explicitly as possible at least helps toward
perceiving also its implications, limitations, and possible alternatives.

Our Brief TherapyBases and Comparisons

Much of the shorter-term treatment that has recently developed in response to the pressure of patient needs and
situational limitations consists essentially of briefer versions of conventional forms of individual or family therapy. The
same basic assumptions are involved, and, correspondingly, the methods used are similar, except for limited adaptations to
the realities of fewer sessions (3, 5, 20). This is expectable, as the usual frameworks naturally offer more restraints to
innovation than encouragement and guidance. Within their terms, new methods are apt to appear strange and unreliable
(15). Consequently, "brief therapy" ordinarily connotes an expedient that may be necessary when a preferred treatment is
not available or is considered not feasiblesince the "best" therapies often require patients equipped with rather
exceptional resources of time, money, intelligence, persistence, and verbal sophistication. The goals of such brief therapy
correspondingly are conceived as limited "first aid"such as relief of some pressing but not fundamental aspect of the
patient's problem, or a supportive holding action until really thorough treatment becomes possible.
We recognize and value the practical and economic advantages for patients and society of shortening treatment. We do
not, however, see our own kind of brief treatment as an expedient, nor is brevity in itself a goal to us, except that we believe
setting time limits on treatment has some positive influence on both therapists and patients. Rather the nature of our
therapy, including its brevity, is primarily a consequence of our premises about the nature and handling of psychiatric
problems.
Our fundamental premise is that regardless of their basic origins and etiologyif, indeed, these can ever be reliably
determinedthe kinds of problems people bring to psychotherapists persist only if they are maintained by ongoing current
behavior of the patient and others with whom he interacts. Correspondingly, if such problem-maintaining behavior is
appropriately changed or eliminated, the problem will be resolved or vanish, regardless of its nature, origin, or duration
(24, 26). Our general principles and specific practices of treatment all relate closely to these two assumptions.
This view, like any other, must be judged by its fruits rather than by its seeds. Yet, a brief consideration of two areas of
shared prior experience and interest that appear to have had major implications for our present joint position may clarify it
and give some due acknowledgement.
Our present brief therapy is visible first as pursuing further two main aspects of family therapy, in which we have all
been extensively involved. A decade-and-a-half ago family therapy began to focus attention on observable behavioral
interaction and its influence, both among family members and between them and the therapist, rather than on long-past
events or inferred mental processes of individuals (10). In line with this, we now see disturbed, deviant, or difficult
behavior in an individual (like behavior generally) as essentially a social phenomenon, occurring as one aspect of a system,
reflecting some dysfunction in that system, and best treated by some appropriate modification of that system. We differ,
however, with those family therapists who consider the dysfunction involved to be necessarily a fundamental aspect of the
system's organization and requiring correspondingly fundamental changes in the system. Instead, we now believe that
apparently minor changes in overt behavior or its verbal labeling often are sufficient to initiate progressive developments.
Further, while we recognize that along with its obvious disadvantages symptomatic behavior usually has some recognizable
advantages or "pay-offs"such as providing leverage in controlling relationshipswe no longer consider these especially
significant as causes of problems or obstacles to change.
Family therapy also has prompted greater activity by therapists. Once family interaction was seen as significant for
problems, it followed that the therapist should aim to change the going system. Extending this, we now see the therapist's
primary task as one of taking deliberate action to alter poorly functioning patterns of interaction as powerfully, effectively,
and efficiently as possible.
On the matter of how the therapist can actively influence behavior effectivelythe strategy and techniques of
changewe are especially indebted to the hypnotic work of Milton Erickson and his closely related psychotherapy. 1 Two
points have been particularly influential. First, although Erickson is much concerned with how overt behavior affects

2
_____________________________________________________________________________________________________________

feelings or states of mind, his moves to change existing behavior usually depend upon implicit or indirect means of
influence. Even when behavior is explicitly discussed, his aim often is not to clarify the "reality" of a situation but to alter
and ameliorate it by some redefinition. Second, both as hypnotist and therapist, Erikson has emphasized the importance of
"accepting what the client offers," and turning this to positive usein ways we will illustrate latereven if what is
"offered" might ordinarily appear as resistance or pathology.
While our present approach thus derives directly from basic family therapy, in part, and from Erickson's work, in part, it
also differs from both. For example, many family therapists attempt to bring about change largely by explicit clarification of
the nature of family behavior and interaction. Such an attempt now seems to us like a family version of promoting "insight,"
in which one tries to make clear to families the covert rules that have guided them; we ordinarily avoid this. Meanwhile, our
conceptualization of problems and treatment appears at least more general and explicit than Erickson's and probably
different in various specific respects.
On the other hand, similarities as well as differences are observable between our treatment approach and other
approaches with which we have had little interaction. For example, within the general field of family therapy, we share with
the crisis-intervention therapy of Pittman, Langsley, and their co-workers (18) beliefs in the importance of situational
change for the onset of problems and of both directive measures and negotiation of conflicts in promoting better functioning
in family systems. Minuchin and Montalvo (16), together with a number of their colleagues at the Philadelphia Child
Guidance Clinic, have increasingly emphasized active intervention aimed at particular re-orderings of family relationship
structure to achieve rapid problem resolution; we often pursue similar aims. Other family therapists than ourselves, notably
Bowen, assign patients homework as part of treatment. Work with families similar to our own is also being developed
abroad, for instance, at the Athenian Institute of Anthropos under Dr. George Vassiliou and at the Istituto per lo Studio
della Famiglia in Milan, under Prof. Dr. Mara Selvini Palazzoli. In addition, the behavior modification school of therapy
involves a number of ideas and interventions rather parallel to ours, although that field still appears to give little attention to
systems of interaction. Furthermore, as noted later, a number of the techniques of intervention we utilize have also been
used and described, though usually in a different conceptual context, by other therapists.
In sum, many particular conceptual and technical elements of our approach are not uniquely ours. We do, however, see
as distinctive the overall system of explicitly stated and integrated ideas and practices that constitute our approach.

MAIN PRINCIPLES OF OUR WORK

1. We are frankly symptom-oriented, in a broad sense. Patients or their family members come with certain complaints and
accepting them for treatment involves a responsibility for relieving these complaints. Also, since deviant symptomatic
behavior and its accompanying vicious circles of reaction and counter-reaction can themselves be so disruptive of system
functioning, we believe that one should not hasten to seek other and deeper roots of pathology. The presenting problem
offers, in one package, what the patient is ready to work on, a concentrated manifestation of whatever is wrong, and a
concrete index of any progress made.
2. We view the problems that people bring to psychotherapists (except, of course, clearly organic psychiatric syndromes)
as situational difficulties between peopleproblems of interaction. Most often this involves the identified paitient and his
family; however, other systems such as a patient's involvement with others in a work situation may be important at times.
3. We regard such problems as primarily an outcome of everyday difficulties, usually involving adaptation to some life
change, that have been mishandled by the parties involved. When ordinary life difficulties are handled badly, unresolved
problems tend increasingly to involve other life activities and relationships in impasses or crises, and symptom formation
results.
4. While fortuitous life difficulties, such as illness, accidents, or loss of a job sometimes appear to initiate the development
of a problem, we see normal transitional steps in family living as the most common and important "everyday difficulties"
that may lead to problems. These transitions include: the change from the voluntary relationship of courtship to the
commitment of marriage, and from this to the less reversible commitment when the first child is born; the sharing of
influence with other authorities required when a child enters school, and with the child himself and his peers in the
adolescent period; the shift from a child-oriented marital relationship back to a two-party system when the children leave
the home, and its intensification at retirement; and return to single life at the death of one spouse. Although most people
manage to handle these transitions at least passably well, they all require major changes in personal relationships that may
readily be mishandled. This view is similar to that of Erickson and Haley (12).
5. We see two main ways by which "problems" are likely to develop: if people treat an ordinary difficulty as a "problem"
or if they treat an ordinary (or worse) difficulty as no problem at allthat is, by either overemphasis or underemphasis of
difficulties in living.
The first appears related to utopian expectations of life. There are countless difficulties which are part and parcel of the
everyday business of living for which no known ideal or ultimate solutions exist. Even when relatively severe, these are
manageable in themselves but can readily become "problems" as a result of a belief that there should or must be an ideal,
ultimate solution for them. For instance, there apparently has been a "generation gap" for the past 5000 years that we know

3
_____________________________________________________________________________________________________________

of, but its difficulties only became greatly exacerbated into a "problem" when many people became convinced that it should
be closed.
Inversely, but equally, "problems" can arise out of the denial of manifest difficultieswhich could be seen as utopian
assertions. For instance, the husband and wife who insist their marriage was made in heaven, or the parents who deny the
existence of any conflicts with their childrenand who may contend that any one seeing any difficulty must be either bad or
madare likely to be laying the foundation for some outbreak of symptomatic behavior.
Two other aspects of this matter need mention. First, over- or under-emphasis of life difficulties is not entirely a matter of
personal or family characteristics; this depends also on more general cultural attitudes and conceptions. While these often
may be helpful in defining and dealing with the common vicissitudes of social life, they can also be unrealistic and provoke
problems. For example, except for the death of a spouse, our own culture characterizes most of the transitions listed earlier
as wonderful steps forward along life's path. Since all of these steps ordinarily involve significant and inescapable
difficulties, such over-optimistic characterization increases the likelihood of problems developingespecially for people
who take what they are told seriously. Second, inappropriate evaluation and handling of difficult situations is often
multiplied by interaction between various parties involved. If two persons have similar inappropriate views, they may
reciprocally reinforce their common error, while if one over-emphasizes a difficulty and another under-emphasizes it,
interaction may lead to increasing polarization and an even more inappropriate stance by each.
6. We assume that once a difficulty begins to be seen as a "problem," the continuation, and often the exacerbation, of this
problem results from the creation of a positive feedback loop, most often centering around those very behaviors of the
individuals in the system that are intended to resolve the difficulty: The original difficulty is met with an attempted
"solution" that intensifies the original difficulty, and so on and on (26).
Consider, for instance, a common pattern between a depressed patient and his family. The more they try to cheer him up
and make him see the positive sides of life, the more depressed the patient is likely to get: "They don't even understand me."
The action meant to alleviate the behavior of the other party aggravates it; the "cure" becomes worse than the original
"disease." Unfortunately, this usually remains unnoted by those involved and even is disbelieved if any one else tries to
point it out.
7. We view long-standing problems or symptoms not as "chronicity" in the usual implication of some basic defect in the
individual or family, nor even that a problem has become "set" over time, but as the persistence of a repetitively poorly
handled difficulty. People with chronic problems have just been struggling inappropriately for longer periods of time. We,
therefore, assume that chronic problems offer as great an opportunity for change as acute problems and that the principal
difference lies in the usually pessimistic expectations of therapists facing a chronic situation.
8. We see the resolution of problems as primarily requiring a substitution of behavior patterns so as to interrupt the
vicious, positive feedback circles. Other less destructive and less distressing behaviors are potentially open to the patient
and involved family members at all times. It is usually impossible, however, for them to change from their rigidly patterned,
traditional, unsuccessful problem-solving behavior to more appropriate behavior on their own initiative. This is especially
likely when such usual behavior is culturally supported, as is often the case: Everyone knows that people should do their
best to encourage and cheer up a loved one who is sad and depressed. Such behavior is both "right" and "logical"but
often it just doesn't work.
9. In contrast, we seek means of promoting beneficial change that works, even if our remedies appear illogical. For
instance, we would be likely to comment on how sad a depressed patient looks and to suggest that there must be some real
and important reason for this. Once given some information on the situation, we might say it is rather strange that he is not
even more depressed. The usual result, paradoxical as it may seem, is that the patient begins to look and sound better.
10. In addition to accepting what the patient offers, and reversing the usual "treatment" that has served to make matters
worse, this simple example also illustrates our concept of "thinking small" by focusing on the symptom presented and
working in a limited way towards its relief.
We contend generally that change can be effected most easily if the goal of change is reasonably small and clearly stated.
Once the patient has experienced a small but definite change in the seemingly monolithic nature of the problem most real to
him, the experience leads to further, self-induced changes in this, and often also, in other areas of his life. That is,
beneficent circles are initiated.
This view may seem insensitive to the "real," "big," or "basic" problems that many therapists and patients expect to be
changed by therapy. Such goals are often vague or unrealistic, however, so that therapy which is very optimistic in concept
easily becomes lengthy and disappointing in actual practice. Views of human problems that are either pessimistic about
change or grandiose about the degree of change needed undermine the therapist's potentially powerful influence for limited
but significant change.
11. Our approach is fundamentally pragmatic. We try to base our conceptions and our interventions on direct observation
in the treatment situation of what is going on in systems of human interaction, how they continue to function in such ways,
and how they may be altered most effectively.
Correspondingly, we avoid the question "Why?" From our standpoint, this question is not relevant, and involvement with
it commonly leads toward concerns about "deeper" underlying causeshistorical, mental, familialof problem behavior

4
_____________________________________________________________________________________________________________

and about "insight" into these.

That is, the question "Why?" tends to promote an individualistic, voluntaristic, and rationalistic conception of human
behavior, rather than one focused on systems of interaction and influence. Moreover, since underlying causes inherently are
inferential rather than observable, concern about them distracts a therapist from close observation of the present problem
and what behavior may be perpetuating it.
On the basis of this general conception of problems and their resolution, which is discussed more fully in Watzlawick,
Weakland, and Fisch (25), we can now describe the overall practical approach and specific techniques that we utilize.

OPERATION OF THE BRIEF THERAPY CENTER

The Brief Therapy Center was established as one of the projects at the Mental Research Institute in January, 1967. Since
the termination of our founding grants, we have continued our work on a somewhat reduced scale on volunteered time.
Some direct operating expenses have been met by donations from patients, although we provide free treatment where
appropriate.
Our working quarters consist of a treatment room and observation room, separated by a one-way viewing screen, with
provision for simultaneously listening to and tape-recording sessions. There is also an intercom phone between the two
rooms. At the outset of our work, a therapist and an official observer were assigned, in rotation, to each case. More
recently, we have been working as an overall team, with several observers of equal status usually present.
Our handling of all cases follows a six-stage schema, although in practice there may be some overlap among these:
1. Introduction to our treatment set-up.
2. Inquiry and definition of the problem.
3. Estimation of behavior maintaining the problem.
4. Setting goals of treatment.
5. Selecting and making behavioral interventions.
6. Termination.
Each of these will now be considered in order.

Introduction to Our Treatment Set-Up

Patients intentionally are accepted with no screening. A first appointment is set by the project secretary whenever an
applicant calls and there is a vacancy in our schedule. No waiting lists are kept; when we have no vacancy, people are
referred elsewhere.
At the first meeting, our secretary has the patient or family fill out a form covering basic demographic data and brings
him or them to the treatment room. The therapist begins by explaining the physical and organizational arrangements,
mentioning the potential advantages for treatment of the recording and observation, and requests written consent to this.
Only two patients have ever declined to proceed on this basis. The therapist also tells the patient at once that we work on a
maximum of ten sessions per case; this helps to set a positive expectation of rapid change.

Definition of the Problem

Since our treatment focus is symptomatic, we want first to get a clear and explicit statement of the presenting complaint.
Therefore, as soon as the therapist has taken a brief record of the referral source and any previous treatment, he asks what
problem has brought the patient to see us. If a patient states a number of complaints, we will ask which is the most
important. In marital or family cases, since viewpoints may differ, although they often are plainly interrelated, we ask each
of the parties involved to state his own main complaint. From the beginning, then, we are following a form of the general
principle, "Start where the patient is at."
Fairly often, the patient will give an adequate answerby which we mean a clear statement referring to concrete
behavior. In many cases, however, the response will leave the presenting problem still in doubt. Further inquiry is then
needed to define more clearly this point of departure for the entire treatment. For example, patients with previous treatment
experience or psychological sophistication are likely, after only the briefest mention of any present behavioral difficulty, to
launch into discussion of presumed underlying matters, especially intrapsychic factors and family history, presenting these
as the "real problem." We then press the question of what particular difficulties in living have brought them to see us now.
To make things more specific, we often ask such questions as "What do you now do because of your problem that you want
to stop doing, or do differently?" and "What would you like to do that your problem interferes with doing now?" Such
inquiries also begin to raise the related question of treatment goals.
Other patients, especially younger ones, may state their complaints in vague terms that lack reference to any concrete
behavior or life situation: "I don't know who I really am"; "We just can't communicate." Such patients can be particularly

5
_____________________________________________________________________________________________________________

difficult initially. We find it important not to accept such statements as appropriate and informative but to continue inquiry
until at least the therapist, if not the patient, can formulate a concrete, behavioral picture of the problemof which such
attachment to vague and often grandiose thinking and talking may itself be a major aspect.

Estimation of Behavior Maintaining the Problem

Our view, as mentioned earlier, is that problem behavior persists only when it is repeatedly reinforced in the course of
social interaction between the patient and other significant people. Usually, moreover, it is just what the patient and these
others are doing in their efforts to deal with the problemoften those attempts at help that appear most "logical" or
unquestionably rightthat is most important in maintaining or exacerbating it.
Once behavior is observed and considered in this light, the way this occurs is often rather obvious: The wife who nags
her husband and hides his bottle in her efforts to save him from his alcohol problem and succeeds only in continually
keeping drinking uppermost in his mind; the forgiving husband who never criticizes his wife until she feels he doesn't care
anything about her, whatever she does, and becomes depressedand he is forgiving of that too; the parents of a child
dissatisfied with school who "encourage" him by talking all the more about how important and great education isinstead
of it being a necessary drag. In other instances, of course, the reinforcements may be more difficult to perceive, either
because they are subtle or complexnon-verbal behaviors, contradictions between statements and actions, different
behaviors by several personsor because even therapists are conditioned to accept cultural standards of logic and
rightness without examining whether things really work that way.
In practice, the therapist first simply asks the patient and any family members present how they have been trying to deal
with the problem. This alone may lead rapidly to a view of what keeps things going badly. If not, the inquiry, aiming always
at concrete behavior, can be pursued at more length and in more detail, but sympatheticallythe therapist's aim is to get
enough information to understand what is happening, for which he needs cooperation, not to confront people with their
mistakes. In addition to what the patient or others state explicitly, it is important to note how they discuss the problem and
its handling, including their interaction. Such inquiry is likely to disclose a number of things that play some part in
maintaining the problem, but working briefly demands choosing priorities. On the basis of observation and experience, one
must judge which behavior seems most crucial.

Setting Goals of Treatment

Setting a goal both acts as a positive suggestion that change is feasible in the time allotted and provides a criterion of
therapeutic accomplishment for therapist and patient. We, therefore, want goals stated clearly in terms of observable,
concrete behavior to minimize any possibility of uncertainty or denial later. If parents bring us a child because he is failing
in school, we ask for an explicit criterion of satisfactory progressbecause we want to avoid subsequent equivocations
such as "He is getting B's now instead of F's, but he isn't really learning enough." Also, we steer toward "thinking small" for
reasons already discussed. Therefore, our usual inquiry is something like "At a minimum, what (change in) behavior would
indicate to you that a definite step forward has been made on your problem?"
Concerning goals especially, however, patients often talk in vague or sweeping terms, despite our efforts to frame the
question in terms of specific behavior. We then try to get more concrete answers by further discussion, clarification, and
presentation of examples of possible goals for consideration. With vague, grandiose, or utopian patients, we have found it
helpful to reverse our field, bringing them down to earth by suggesting goals that are too far out even for them. This again
involves accepting what the patient offers, and even enlarging on this, in order to change it. For example, a student who was
already in his mid-20's and was still being supported by a working mother told us he was studying "philosophical
anthropology" in order to bring the light of India and China to bear on the West. He also, however, mentioned some interest
in attending a well-known school of Indian music. It was then pointed out to him that this represented a rather limited aim
compared to his concern to unite the spirituality of India with the practical communism of China and use both to reconstruct
Western society. He then said that, since he was not doing well in his studies and was short of money, if he could secure a
scholarship and really learn Indian music, this would be quite enough accomplishment for the present.
We usually are able, directly or indirectly, to obtain a stated goal that appears sufficiently explicit and appropriate to the
problem. In some cases, however, we have not been able to do so. Either the patient persisted in stating only vague,
untestable goals, or, rarely, the patient stated and stuck to an explicit goal which we judged inappropriate to his problem.
Then we do not dispute what the patient insists on but privately set our own goal for the case by joint staff discussion of
what sort of behavior would best exemplify positive change for the particular patient and problem. In fact, some such
discussion occurs for all cases; at the least, the staff must always judge whether the patient's statement of his goal is
adequate. Also, there is always staff discussion of intermediate behavioral goals; how does the patientor his family
membersneed to behave so that the specific goal of treatment will follow?2

6
_____________________________________________________________________________________________________________

Our aim is to have a definite goal established by the second session, but gathering and digesting the information needed
for this sometimes takes longer. Occasionally, we may revise the original goal in the course of treatment or add a secondary
goal.

Selecting and Making Interventions

Once we have formed a picture of current behavior central to the problem and estimated what different behavior would
lead to the specific goal selected, the task is one of intervening to promote such change. This stage must be discussed at
some length, since it ordinarily constitutes the largest, most varied, and probably most unusual part of our treatment.
Change and "insight." We have already stated that our aim is to produce behavior change and that we do not see
working toward insight, at either an individual or a family level, as of much use in this. In fact, working toward insight can
even be counter-productive. Simple, practical-minded patients are often put off by this, since they want action and results,
while more intellectually minded patients are likely to welcome such an approach but use it to delay or defeat any change in
actual behavior. However, in addition to suggesting or prescribing changes in overt behavior, we do utilize interpretations.
Our aim, though, is simply the useful relabeling of behavior. Patients often interpret their own behavior, or that of others, in
ways that make for continuing difficulties. If we can only redefine the meaning or implications attributed to the behavior,
this itself may have a powerful effect on attitudes, responses and relationships. Such interpretation might look like an
attempt to impart insight, but it is not. Using interpretation to promote insight implies that truth can helpfully be disclosed
and recognized. This is not our aim or our belief. Rather, our view is that redefining behavior labeled "hostile" as
"concerned interest," for example, may be therapeutically useful whether or not either label is "true," and that such truth can
never be firmly established. All that is observable is that some labels provoke difficulties, while others, achievable by
redefinition, promote adjustment and harmonybut this is enough.
Such relabeling may be especially important with rigid patients. It does not require overt behavior change, and it may
even be accomplished without the need for any active cooperation by the patient or any family member. If the therapist's
redefinition of an action or situation is not openly challengedwhich can usually be arrangedthen the meaning and
effects of that behavior have already been altered.
Use of idiosyncratic characteristics and motivation. We attempt early in treatment to determine what approach would
appeal most to the particular patientto observe "where he lives" and meet this need, whether it is to believe in the
magical, to defeat the expert, to be a caretaker of someone, to face a challenge, or whatever. Since the consequences of any
such characteristic depend greatly on the situation in which it operates and how this is defined, we see these characteristics
of different individuals not as obstacles or deficiencies, but as potential levers for useful interventions by the therapist.
For example, certain patients appear inclined toward defeating therapists, despite their request for help. This may be
indicated by a history of unsuccessful treatment, repeated failure to understand explanations or carry out instructions, and so
on. In such cases, the easiest and most effective course may be for the therapist to insist that the patient cannot possibly
resolve his problem and that treatment can at most help him to endure it better. The patient is then likely to defeat this
stance by improving.
A middle-aged widow first came to us with a complaint about the behavior of her 18-year-old son: delinquency, school
failures, anger, and threatened violence toward her. She stated this was her only problem, although she also mentioned that
she was an epileptic and was unable to use her right arm as a result of a work injury. Both mother and son had had about
two years of previous therapy. We first suggested directly that her son was acting like a difficult, provoking, overgrown kid
and, accordingly, she might gain by handling him more firmly in a few simple ways. She quickly thwarted such suggestions
by increasing claims of helplessness: Now the epilepsy was emphasized; there was trouble with the other arm, too; a
hysterectomy and appendectomy were also reported, along with childhood rheumatic fever, bleeding gums, troubles with
her former husband and with her mother-in-law, constant worsening financial crises, and much more. In short, she was
already a woman carrying on bravely amidst a sea of troubles that would have totally swamped anyone else; how could we
ask her to do more yet? We then changed our approach to utilize this characteristic opposition. We began to insist to her
that she was being unduly optimistic, was minimizing her troubles in an unrealistic way, and was not recognizing that the
future very probably held even greater disasters for her, both individually and in terms of her son's behavior. It took some
doing to surpass her own pessimistic line, but once we were able to do so, she began to improve. She started to oppose our
pessimismwhich she could only do by claiming and proving that she was not that sick and helplessand to take a much
more assertive attitude with her son, to which he responded well.
Directed behavior change. One of our main stated aims is to change overt behaviorto get people to stop doing things
that maintain the problem and to do others that will lead toward the goal of treatment. While we are willing to issue
authoritative directions, we find compliant patients rather rare. After all, most patients have already been exposed to lots of
advice. If it was good, they must have some difficulty about profiting from advice; if it was bad, some preparation is needed
for them to respond to quite different advice. Moreover, again, it is often just that behavior that seems most logical to

7
_____________________________________________________________________________________________________________

people that is perpetuating their problems. They then need special help to do what will seem illogical and mistaken. When
sitting on a nervous horse, it is not easy to follow the instructor's orders to let go of the reins. One knows the horse will run
away, even though it is really the pull on the reins that is making him jump.
Behavioral instructions therefore are more effective when carefully framed and made indirect, implicit, or apparently
insignificant. When requesting changes, it is helpful to minimize either the matter or the manner of the request. We will
suggest a change rather than order it. If the patient still appears reluctant, we will back off further. We may then suggest it is
too early to do that thing; the patient might think about it but be sure not to take any action yet. When we do request
particular actions, we may ask that they be done once or twice at most before we meet again. We may request only actions
that will appear minor to the patient, although in our view they represent the first in a series of steps, or involve a
microcosm of the central difficulty. For example, a patient who avoids making any demands of others in his personal
relationships may be assigned the task of asking for one gallon of gasoline at a service station, specifically requesting each
of the usual free services, and offering a twenty-dollar bill in payment [sic].
This example also illustrates our use of "homework" assignments to be carried out between sessions. Homework of
various kinds is regularly employed, both to utilize time more fully and to promote positive change where it counts most, in
real life outside the treatment room.
Paradoxical instructions. Most generally, paradoxical instruction involves prescribing behavior that appears in
opposition to the goals being sought, in order actually to move toward them. This may be seen as an inverse to pursuing
"logical" courses that lead only to more trouble. Such instructions probably constitute the most important single class of
interventions in our treatment. This technique is not new; aspects and examples of it have been described by Frankl (8, 9),
Haley (11), Newton (17) and Watzlawick, et al. (24). We have simply related this technique to our overall approach and
elaborated on its use.
Paradoxical instruction is used most frequently in the form of case-specific "symptom prescription," the apparent
encouragement of symptomatic or other undesirable behavior in order to lessen such behavior or bring it under control. For
example, a patient who complains of a circumscribed, physical symptomheadache, insomnia, nervous mannerisms, or
whatevermay be told that during the coming week, usually for specified periods, he should make every effort to increase
the symptom. A motivating explanation usually is given, e.g., that if he can succeed in making it worse, he will at least
suffer less from a feeling of helpless lack of control. Acting on such a prescription usually results in a decrease of the
symptomwhich is desirable. But even if the patient makes the symptom increase, this too is good. He has followed the
therapist's instruction, and the result has shown that the apparently unchangeable problem can change. Patients often
present therapists with impossible-looking problems, to which every possible response seems a poor one. It is comforting,
in turn, to be able to offer the patient a "therapeutic double bind" (4), which promotes progress no matter which alternative
response he makes.
The same approach applies equally to problems of interaction. When a schizophrenic son used bizarre, verbal behavior
to paralyze appropriate action by his parents, we suggested that when he needed to defend himself against the parents'
demands, he could intimidate them by acting crazy. Since this instruction was given in the parents' presence, there were two
paradoxical positive effects: the son decreased his bizarreness and the parents became less anxious and paralyzed by any
such behavior.
Not infrequently, colleagues find it hard to believe that patients will really accept such outlandish prescriptions, but they
usually do so readily. In the first place, the therapist occupies a position of advice-giving expert. Second, he takes care to
frame his prescriptions in a way most likely to be accepted, from giving a rationale appropriate to the particular patient to
refusing any rationale on the grounds that the patient needs to discover somethings quite unanticipated. Third, we often are
really just asking the patient to do things they already are doing, only on a different basis.
We may also encourage patients to use similar paradoxes themselves, particularly with spouses or children. Thus, a
parent concerned about her child's poor school homework (but who probably was covertly discouraging him) was asked to
teach the child more self-reliance by offering incorrect answers to the problems he was asking help in solving.
Paradoxical instructions at a more general level are often used also. For example, in direct contrast to our name and
ten-session limit, we almost routinely stress "going slow" to our patients at the outset of treatment and, later, by greeting a
patient's report of improvement with a worried look and the statement, "I think things are moving a bit too fast." We also do
the same thing more implicitly, by our emphasis on minimal goals, or by pointing out possible disadvantages of
improvement to patients, "You would like to do much better at work, but are you prepared to handle the problem of envy by
your colleagues?" Such warnings paradoxically promote rapid improvement, apparently by reducing any anxiety about
change and increasing the patient's desire to get on with things to counteract the therapist's apparent overcautiousness.
On the same principle, when a patient shows unusually rapid or dramatic improvement, after acknowledging this change
we may prescribe a relapse, on the rationale that it further increases control: "Now you have managed to turn the symptom
off. If you can manage to turn it back on during this next week, you will have achieved even more control over it." This

8
_____________________________________________________________________________________________________________

intervention, similar to Rosen's "re-enacting the psychosis" (18) and related techniques of Erickson, anticipates that in some
patients improvement may increase apprehension about change and meets this danger by paradoxically redefining any
relapse that might occur as a step forward rather than backward.
Since we as therapists are by definition experts, giving authoritative instructions on both thinking and acting, another
pervasive element of paradox is created by the fact that ordinarily we do so only tentatively, by suggestions or questions
rather than direct orders, and often adopt a "one-down" position of apparent ignorance or confusion. We find that patients,
like other people, accept and follow advice more readily when we avoid "coming on strong."
Utilization of interpersonal influence. Although many of our treatment sessions include directly only one therapist and
one patient, we consider and utilize more extended interpersonal relationships constantly in our work. First, even when we
see only the "identified patient," we conceive the problem in terms of some system of relationships and
problem-maintaining behavior involving his family, his friends, or his work situation. Therefore, we believe that any
interventions made with the patient must also take their probable consequences for others into account. Equally, however,
useful interventions may be made at any point in the system, and frequently it appears more effective to focus our efforts on
someone other than the identified patient. Where a child is the locus of the presenting problem, we very commonly see the
whole family only once or twice. After this we see the parents only and work with them on modifying their handling of the
child or their own interaction. With couples also, we may see the spouses separately for the most part, often spending more
time with the one seen by them as "normal." Our point is that effective intervention anywhere in a system produces changes
throughout, but according to what the situation offers, one person or another may be more accessible to us, more open to
influence, or a better lever for change in the system.
Second, the therapist and the observers also constitute a system of relationships that is frequently used to facilitate
treatment. With patients who find it difficult to accept advice directly from a real live person, an observer may make
comments to the therapist over the intercom phone to be relayed to the patient from this unseen and presumably objective
authority. When a patient tends to disagree constantly, an observer may enter and criticize the therapist for his "poor
understanding" of the case, forming an apparent alliance with the patient. The observer can then often successfully convey
re-phrased versions of what the therapist was offering originally. With patients who alternate between two different stances,
two members of the treatment team may agree, separately, with the two positions. Then, whatever course the patient takes
next he is going along with a therapist's interpretation, and further suggestions can be given and accepted more
successfully. Such therapist-observer interaction strategies can bring about change rapidly even with supposedly "difficult"
patients.3
As may be evident, all of these techniques of intervention are means toward maximizing the range and power of the
therapist's influence. Some will certainly see, and perhaps reject, such interventions as manipulative. Rather than arguing
over this, we will simply state our basic view. First, influence is an inherent element in all human contact. Second, the
therapist's functioning necessarily includes this fact of life, but goes much further; professionally he is a specialist at
influence. People come to a therapist because they are not satisfied with some aspect of their living, have been unable to
change it, and are seeking help in this. In taking any case, therefore, the therapist accepts the assignment of influencing
people's behavior, feelings, or ideas toward desirable ends. Accordingly, third, the primary responsibility of the therapist is
to seek out and apply appropriate and effective means of influence. Of course, this includes taking full account of the
patient's stated and observed situation and aims. Given these, though, the therapist still must make choices of what to say
and do, and equally what not to say and do. This inherent responsibility cannot be escaped by following some standard
method of treatment regardless of its results, by simply following the patient's lead, or even by following a moral ideal of
always being straightforward and open with the patient. Such courses, even if possible, themselves represent strategic
choices. To us, the most fundamental point is whether the therapist attempts to deny the necessity of such choices to
himself, not what he tells the patient about them. We believe the better course is to recognize this necessity, to try whatever
means of influence are judged most promising in the circumstances, and to accept responsibility for the consequences.
Termination. Whether cases run the limit of ten sessions or goals are achieved sooner, we usually briefly review the
course of treatment with the patient, pointing out any apparent gainsgiving the patient maximum credit for this
achievementand noting any matters unresolved. We also remark on the probable future beyond termination, ordinarily in
connection with reminding patients that we will be contacting them for a follow-up interview in about three months. This
discussion usually embodies positive suggestions about further improvement. We may remind patients that our treatment
was not intended to achieve final solutions, but an initial breakthrough on which they themselves can build further. In a
minority of cases, howeverparticularly with negativistic patients, ones who have difficulty acknowledging help from
anyone, or those fond of challengeswe may take an opposite tack, minimizing any positive results of treatment and
expressing skepticism about any progress in the future. In both instances, our aim is the same, to extend our therapeutic
influence beyond the period of actual contact.
In some cases, we encounter patients who make progress but seem unsure of this and concerned about termination. We
often meet this problem by means of terminating without termination. That is, we say we think enough has been

9
_____________________________________________________________________________________________________________

accomplished to terminate, but this is not certain; it can really be judged only by how actual life experience goes over a
period of time. Therefore, we propose to halt treatment, but to keep any remainder of the ten sessions "in the bank,"
available to draw on if the patient should encounter some special difficulty later. Usually, the patient then departs more at
ease and does not call upon us further.

EVALUATION AND RESULTS

If psychotherapy is to be taken seriously as treatment, not just an interesting exploratory or expressive experience, its
effectiveness must be reliably evaluated. But this is far from easy, and rather commonly therapists offer only general clinical
impressions of their results, with no follow-up of cases after termination, while researchers present ideal study designs that
seldom get implemented.
We certainly cannot claim to have resolved this problem fully, even though we have been concerned with systematic
evaluation of results from the outset of our work. Our method of evaluation still involves some clinical judgments and
occasional ambiguities, despite efforts to minimize these. Until very recently, we have not had the resources needed to
repeat our short-term follow-ups systematically after longer periods. And our evaluation plan is apt to seem overly simple
in comparison with such comprehensive schemes as that of Fiske, et al. (6). At most, we can claim only that our method of
evaluation is simple, avoiding dependence upon either elaborate manipulation and interpretation of masses of detailed data
or elaborate theoretical inference; that it is reasonably systematic and practicable; and most important, that it is consonant
with our overall approach to problems and treatment.
We see the essential task of evaluation as systematic comparison of what treatment proposes to do and its observable
results. Our treatment aim is to change patients' behavior in specific respects, in order to resolve the main presenting
complaint. Give the brevity of our work, the past refractoriness of most of the problems presented, and our frequent
observation of behavior change immediately following particular interventions, we feel fairly safe in crediting observed
changes to our treatment. Our evaluation then depends on answers to the two questions: Has behavior changed as planned?
Has the complaint been relieved?
In our follow-up, the interviewer, who has not participated in the treatment, first inquires whether the specified treatment
goal has been met. For instance, "Are you still living with your mother, or are you living in your own quarters now?" Next,
the patient is asked the current status of the main complaint. This is supplemented by inquiring whether any further therapy
has been sought since terminating with us. The patient is also asked whether any improvements have occurred in areas not
specifically dealt with in treatment. Finally, to check on the supposed danger of symptom substitution, the patient is
routinely asked if any new problems have appeared.
Ideally, such evaluation would divide our cases into two neat piles: successes in which our goal of behavior change was
met and the patient's problem completely resolved, and failures in both respects. In reality, our treatment is not perfect;
while results in these terms are clear for a majority of cases, several sources of less clear-cut outcomes remain: (a) Fairly
often we have had cases in which our goal was reached or approached and considerable improvement was evident, but
complete resolution of the presenting problem or problems was not attained. (b) Occasionally we have failed to formulate a
goal explicit and concrete enough to check on its achievement with certainty. (c) In a very few cases, achievement of the
planned goal and reported relief of the problem have been inversely relatedhitting our target of change did not lead to
relief, or we somehow got results in spite of missing our specific target.
In terms of our basic principles, all such mixed cases must be considered as failures of either conception or execution
that demand further study. In the patients' terms, on the other hand, some of these cases have been completely successful,
and many others represent quite significant progress. For the more limited and immediate purpose of evaluating the general
utility of our approach, therefore, we have classified our cases into three groups according to practical results, recognizing
that these correlate generally but not completely with achievement of our specific goals of behavior change. These groups
represent: (a) complete relief of the presenting complaint; (b) clear and considerable, but not complete, relief of the
complaint; and (c) little or no such change. For simplicity, the one case in which things were worse after treatment is
included in the third group. We have not broken down our sample into sub-groups based on common diagnosis, since the
conventional system of diagnostic categories and our conception of problems and their treatment are based on different
assumptions and the nature of the presenting problem has appeared to make little difference for our rate of success or
failure. It should also be noted that this evaluation refers directly only to the major presenting complaint. However, in none
of our cases in which this complaint was resolved was there any report of new problems arising, and in many of these
improvements in additional areas were reported. On this basis, then, our overall results for 97 cases, involving an average
of 7.0 sessions, are:

Success 39 cases 40 per cent

Significant improvement 31 cases 32 per cent

10
_____________________________________________________________________________________________________________

Failure 27 cases 28 per cent

These results appear generally comparable to those reported for various forms of longer-term treatment.

CONCLUSION: IMPLICATIONS
In this paper we have set forth a particular conception of the nature of psychiatric problems, described a corresponding
brief treatment approach and techniques, and presented some results of their application. Clearly, further clinical research
should be done, as important problems obviously remain; goals are still difficult to set in certain types of cases, the choice
of interventions has not been systematized, evaluation is not perfected. Concurrently, though, there should also be more
thinking about the broader significance of these ideas and methods. Our results already give considerable evidence for the
usefulness of our general conception of human problems and their practical handling. Since this is both quite different from
more common views and potentially widely relevant, we will conclude with a tentative consideration of some broad
implications of our work.
The most immediate and evident potential of our work is for more effective use of existing psychiatric facilities and
personnel. This could include reduction in the usual length of treatment and a corresponding increase in the number of
patients treated, with no sacrifice of effectiveness. In fact, our approach gives promise of more than ordinary effectiveness
with a variety of common but refractory problems, such as character disorders, marital difficulties, psychoses, and chronic
problems generally. Further, it is not restricted to highly educated and articulate middle-class patients but is applicable to
patients of whatever class and educational background.
In addition, our approach is relatively clear and simple. It might therefore be feasible to teach its effective use to
considerable numbers of lay therapists. Even if some continuing supervision from professionals should be necessary, the
combination of brief treatment and many therapists thus made possible could help greatly in meeting present needs for
psychological help. Although this kind of development would have little to offer private practice, it could be significant for
the work of overburdened social agencies.
Taking a wider view, it is also important that our model sees behavioral difficulties "all under one roof" in two respects.
First, our model interrelates individual behavior and its social context instead of dividing themnot only within the family,
but potentially at all levels of social organization. Second, this framework helps to identify continuities, similarities, and
interrelations between normal everyday problems, psychiatric problems of deviant individual behavior, and many sorts of
socially problematic behavior, such as crime, social isolation and anomie, and certain aspects of failure and poverty. At
present, social agencies attempting to deal with such problems at the individual or family level are characterized by marked
conceptual and organizational divisionsbetween psychological vs. sociological, supportive vs. disciplinary orientations,
and more specifically, in the division of problems into many categories that are presumed to be distinct and
discretereminiscent of the "syndromes" of conventional psychiatry. At best, this results in discontinuity; ineffective,
partial approaches; or reduplication of efforts. At worst, it appears increasingly likely that such divisions themselves may
function to reinforce inappropriate attempts at solution of many kinds of problems, as suggested by Auerswald (1) and
Hoffman and Long (14). Our work thus suggests a need and a potential basis for a more unified and effective organization
of social services.
Finally, our work has still broader implications that deserve explicit recognition, even though any implementation
necessarily would be a very long-range and difficult problem. Our theoretical viewpoint is focused on the ways in which
problems of behavior and their resolution are related to social interaction. Such problems occur not only with individuals
and families, but also at every wider level of social organization and functioning. We can already discern two kinds of
parallels between problems met in our clinical work and larger social problems. Problems may be reduplicated widely, as
when concern about differences between parents and children becomes, in the large, "the generation gap problem." And
conflicts between groupswhether these groups are economic, racial, or politicalmay parallel those seen between
individuals. Our work, like much recent social history, suggests very strongly that ordinary, "common-sense" ways of
dealing with such problems often fail, and, indeed, often exacerbate the difficulty. Correspondingly, some of our uncommon
ideas and techniques for problem-resolution might eventually be adapted for application to such wider spheres of human
behavior.

REFERENCES
1. Auerswald, E., "Interdisciplinary vs. Ecological Approach," Fam. Proc., 7, 202-215, 1968.
2. Barten, H. (Ed.), Brief Therapies, New York, Behavioral Publications, 1971.
3. Barten, H. and Barten, S. (Eds.), Children and Their Parents in Brief Therapy, New York, Behavioral
Publications, 1972.

11
_____________________________________________________________________________________________________________

4. Bateson, G., Jackson, D., Haley, J. and Weakland, J., "Towards a Theory of Schizophrenia," Behav. Sci., 1,
251-2645, 1956.
5. Bellak, L. and Small, L., Emergency Psychotherapy and Brief Psychotherapy, New York, Grune and Stratton,
1965.
6. Fiske, D., Hunt, H., Luborsky, L., Orne, M., Parloff, M., Reiser, M. and Tuma, A., "Planning of Research on
Effectiveness of Psychotherapy," Arch. Gen. Psychiat., 22, 22-32, 1970.
7. Frank, J., Persuasion and Healing, Baltimore, Johns Hopkins Press, 1961.
8. Frankl, V., The Doctor and the Soul, New York, Alfred A. Knopf, 1957.
9. Frankl, V., "Paradoxical Interventions," Amer. J. Psychother., 14, 520-535, 1960.
10. Jackson, D. and Weakland, J., "Conjoint Family Therapy: Some Considerations on Theory, Technique, and
Results," Psychiatry, Supplement to, 24:2, 30-45, 1961.
11. Haley, J., Strategies of Psychotherapy, New York, Grune and Stratton, 1963.
12. Haley, J., Uncommon Therapy: The Psychiatric Techniques of Milton H. Erickson, M.D., New York, W. W.
Norton, 1973.
13. Haley, J. (Ed.), Advanced Techniques of Hypnosis and Therapy: Selected Papers of Milton H. Erickson, M.D.,
New York, Grune and Stratton, 1969.
14. Hoffman, L. and Long, L., "A Systems Dilemma," Fam. Proc., 8, 211-234, 1969.
15. Krohn, A., "Beyond Interpretation," (A review of M.D. Nelson, et al., Roles and Paradigms in Psychotherapy),
Contemporary Psychology, 16, 380-382, 1971.
16. Minuchin, S. and Montalvo, B., "Techniques for Working with Disorganized Low Socioeconomic Families,"
Amer. J. Orthopsychiat., 37, 880-887, 1967.
17. Newton, J., "Considerations for the Psychotherapeutic Technique of Symptom Scheduling," Psychotherapy:
Theory, Research and Practice, 5, 95-103, 1968.
18. Pittman, F. S., Langsley, D. G., Flomenhaft, K., De Young, C. D., Machotka, P. and Kaplan, D. M., "Therapy
Techniques of the Family Treatment Unit," pp. 259-271 in Haley, J. (Ed.), Changing Families: A Family Therapy
Reader, New York, Grune and Stratton, 1971.
19. Rosen, J., Direct Analysis, New York, Grune and Stratton, 1953.
20. Rosenthal, A., Report on brief therapy research to the Clinical Symposium, Department of Psychiatry, Stanford
University Medical Center, November 25, 1970.
21. Rosenthal, R., Experimenter Effects in Behavioral Research, New York, Appleton-Century-Crofts, 1966.
22. Saizman, L., "Reply to the Critics," Int. J. Psychiat., 6, 473-478, 1968.
23. Spiegel, H., "Is Symptom Removal Dangerous?", Amer. J. Psychiat., 123, 1279-1283, 1967.
24. Watzlawick, P., Beavin, J. and Jackson, D., Pragmatics of Human Communication, New York, W. W. Norton,
1967.
25. Watzlawick, P., Weakland, J. and Fisch, R., Change: Principles of Problem Formation and Problem Resolution,
New York, W. W. Norton, 1974.
26. Wender, H., "The Role of Deviation-Amplifying Feedback in the Origin and Perpetuation of Behavior,"
Psychiatry, 31, 317-324, 1968.

Reprint requests should be addressed to: John H. Weakland, Brief Therapy Center, Mental Research Institute, 555
Middlefield Road, Palo Alto, California 94301.

1The work of Jay Haley (11, 12, 13) has been valuable in making Erickson's principles and practices more explicit, as well as in

providing additional ideas from Haley's own work in family therapy and brief treatment.

2Our schedule is arranged to allow for one half-hour after each session for staff discussion and planning of goals, specific
interventions to use, and so on. In addition, new cases and general issues are considered at more length in separate, weekly staff
meetings.

3Team work facilitates such interventions but actually is seldom essential. A single therapist who is flexible and not unduly

concerned about being correct and consistent can also utilize similar techniquesfor example, by stating two different positions
himself.

12