Chapter 4: Strategic and Systemic Models

Strategic Family Therapy

Overview

The Strategic family therapy models were developed in the 1950s. They arose from two primary sources: first, Gregory Bateson and the Palo Alto Group, who had applied the science of cybernetics to family communication patterns, and second, Milton Erickson, who developed revolutionary paradoxical interventions that capitalized on people’s natural reluctance to change to bring about rapid changes in psychiatric symptoms.

The Palo Alto Group. Gregory Bateson was joined in 1953, first by Jay Haley and John Weakland and later by William Fry. In 1954, Bateson received a grant from the Macy Foundation to study schizophrenia and was then joined by Don Jackson. The group studied the family communication patterns of people diagnosed with schizophrenia to try to determine the origin of the symptoms. Guiding their work were concepts derived from cybernetics, the study of how information-processing systems are controlled by feedback loops. The group viewed families as information processing systems and applied cybernetics concepts to patterns of interactions. They “assumed that psychotic behavior in one member of a family might make sense in the context of pathological family communication” (Nichols & Schwartz, 1998, p. 28), and hypothesized that a complex communication pattern called the double bind (see below) might account for psychotic symptoms in a family member. The proposed explanation was intriguing but controversial since it challenged the prevailing biological “disease” theory. Although the researchers did find disordered communication patterns in these families, no definitive evidence demonstrates that schizophrenic symptoms are the result thereof, and the biological model continues to dominate today.

Double Bind. The term double bind has been commonly misused to simply describe a contradictory message, but the Palo Alto group was referring to interactions that are more complex. There are six characteristics of a double bind (Nichols & Schwartz, 1998):

For example, a father demands that his son engage in a nightly discussion at the dinner table. But when the child attempts to participate, the father is irritated that his dinner is interrupted. The father is then critical of his son’s lack of conversation. The son is caught in a bind since both his attempt to talk and his silence are punished. For the child, the meaning of communication becomes unclear. Over time, he develops a disordered style of communication that the Palo Alto group associated with schizophrenia.

During that same period, Milton Erickson proposed radical new ways to change “psychiatric” symptoms and problems. The prevailing theoretical assumption at the time was that symptoms stemmed from deep psychological problems and that “curing” the problem required that patients gain insight into the unconscious impulses governing their behavior. By contrast, Erickson focused on the specific symptoms and problems presented by the patient. He believed, first, that people had the ability to solve their own problems if they could be induced to try new behaviors; second, that change could be swift rather than a long arduous process; and finally, that the patient’s own natural resistance to change could, ironically, be used to bring about change. As a hypnotherapist, he developed many ingenious techniques for “getting people to do something different in the context of the old behavior, or to do the old behavior in a new context” (Nichols & Schwartz, 1998, p. 358).

Erickson was masterful in his use of therapeutic paradox. In this context, paradox refers to an intervention that appears to contradict therapeutic goals but is designed to promote change by mobilizing resistance. Such interventions are based on the notion that families experiencing symptoms or problems find it difficult, or are naturally resistant to, instituting changes. In those cases, it is sometimes more useful either to forbid them to change or ask them to change in ways that seem to run counter to the desired goals. The therapist is counting on the family members’ rebelling against the directive, and as they do, the desired result is achieved. In a famous example—perhaps fictional, but illustrative nonetheless—a farmer attempts to push his cow into the barn. The cow naturally resists by pushing back against the farmer. The farmer then is instructed to pull the cow backward by the tail away from the barn. The cow resists again by pulling against the farmer, but this time the cow’s resistance lands her in the barn.

Strategic therapy models combine the concepts of the Palo Alto group and Erickson. The defining characteristics of these models of family therapy are:

The models primarily associated with strategic therapy are the MRI Brief Therapy and the Haley–Madanes strategic models. They are presented below, together with Bandler and Grinder’s model—neuro-linguistic programming (NLP).

Mental Research Institute (MRI)

The earliest strategic model came from the work at the Mental Research Institute (MRI) founded in 1959 in Palo Alto by Bateson’s colleague, Don Jackson, who was joined by Jay Haley, Virginia Satir, John Weakland, Paul Watzlawick, Arthur Bodin, and Janet Beavin (Nichols & Schwartz, 1998). They were interested in family communication patterns and feedback loop mechanisms (see below). The MRI group published many articles in the 1960s and 1970s and started one of the first formal training programs in family therapy (Nichols & Schwartz, 1998). In 1967, Haley left MRI for the Philadelphia Child Guidance Clinic.

Brief Therapy Center. In 1967, the Brief Therapy Center opened at MRI. As with all strategic therapies, the goal of treatment is to change the presenting complaint rather than to interpret the interactions to the family or to explore the past. The therapist first assesses the cycle of problematic interactions, then breaks the cycle by using either straightforward or paradoxical directives (Piercy et al., 1996). In this model, the therapist designs or selects a task or directive in order to solve the problem. Thus, the therapist assumes full responsibility for the success or failure of treatment.

Theory of Normal Development and Dysfunction

MRI therapists do not speculate about normative patterns of development or use specific criteria to measure the health of a family. The model is more focused on techniques for change than on theoretical constructs (Piercy et al., 1996). MRI therapists are not concerned with changing the organization of a family (e.g., its hierarchy or power structure). Rather, they focus on faulty cycles of interaction that are usually set into motion by misguided attempts to solve problems. Instead of solving the problem, the family’s attempts may maintain or even worsen it. Problems are not viewed as having linear causes; rather, a problem behavior is viewed as just one point in a repetitive pattern. Causality is circular.

MRI therapists are guided by the principles derived from cybernetics. Cybernetics is the study of how information-processing systems are self-correcting, controlled by feedback loops. Feedback loops are the mechanisms or cycles of interactions through which information is returned to the system and exerts an influence on it. There are both negative and positive feedback loops.

Negative Feedback Loops are ways in which families correct a deviation in their functioning in order to return it to a previous state of homeostasis.

Positive Feedback Loops (Deviation Amplification) arise as a family attempts to introduce new information into the system. This can occur as a part of the growth process or increasing complexity. However, positive feedback loops are also assumed to be responsible for the development of problems in families as they attempt solutions that inadvertently worsen or maintain the problem.

For example, if a child misbehaves, i.e., deviates from the norm, (the family problem) because he is jealous of a new sibling and the father responds with harsh or punitive behavior (an attempted solution), it confirms the child’s belief that he is loved less, leading to an escalation of the child’s behavior (the deviation is amplified). MRI interventions would be aimed at changing the pattern of interaction so that the father could help the child regulate his behavior and convey that he is not loved less.

Assessment and Treatment

Assessment consists of determining the feedback loops that govern the faulty behavior patterns by observing repetitive patterns of family interactions. Treatment is usually limited to 10 sessions, which sets up a “powerful expectation for change” (Nichols & Schwartz, 1998, p. 368). The changes that occur through treatment are classified as first-order change or second-order change.

First-Order Change. Family patterns of interaction or sequences are altered at the behavioral level only.

Second-Order Change. The family rules or underlying beliefs or premises that govern family members’ behavior or promote specific reactions are altered. In the above example, two of the father’s beliefs (that children should never show disrespect and that the child’s behavior is disrespectful) may need to be changed. Family rules may be changed by the technique of reframing (see below), helping the father reinterpret the child’s behavior as reflecting his unhappiness rather than as being disrespectful.

Treatment follows a six-step procedure (outlined by Nichols & Schwartz, 1998, pp. 367–368):

b. Exploring previous attempts to solve the problem. It is helpful to know what solutions the family has already tried for several reasons. The behaviors associated with attempts at solving the problem may be maintaining the problem. Knowing about these attempts helps the therapist avoid strategies that repeat the family’s efforts and point to other strategies. There are three general types of solutions the family may have tried, and each suggests a specific intervention strategy.

The family might have:

Symptom Prescription. The family is requested to continue to perform or even expand the symptom. The intervention may be compliance-based if the therapist wants the family to do as suggested or defiance-based when he/she wants the family to defy the directive.

Restraining Techniques. Family members are warned of the dangers of change, restrained from trying to change, or asked to change slowly. The ‘restraint of change’ technique is used when the family seems ambivalent about changing. The therapist aligns with the side of the ambivalence that resists change so that the family will align with the side that wishes to change.

Positioning. The therapist amplifies or exaggerates the family’s explanation of the problem to a point where the family will disagree.

♦♦♦♦♦

Weakland, J. H., Fisch, R., Watzlawick, P., & Bodin, A. M. (1974). Brief therapy: Focused problem resolution. Family Process, 13(2), 141–167.

In this classic article, the authors describe The Brief Therapy Center of MRI in Palo Alto and their model of therapy. Brief therapists view dysfunctional behavior as a social phenomenon that occurs as just one part of a system. The MRI approach does not consider that “payoffs” or the advantages of symptomatic behavior contribute significantly to problems or hinder change.

The methods of therapy draw from the work of Milton Erickson in two ways. First, Erickson’s goal was to modify a problem by redefining rather than clarifying it. Second, Erickson designed a creative strategy based on a client’s own starting point. Brief Therapy also draws from, among others, the work of Jay Haley. The main principles of Brief Therapy are as follows:

The Brief Therapy approach is pragmatic. Interventions are based on direct observation in the session about how a behavior functions. Understanding why a behavior occurs (insight) is not a goal of therapy. In fact, attending to such inferences may detract from observing the system.

The Brief Therapy Center involves a team consisting of observers and a therapist. The team uses a room with a one-way mirror for observation, a telephone connecting the observers with the therapist, and equipment to record the sessions. The therapist may offer suggestions to the family, or the team of observers may intervene. These therapist–observer interventions have been found helpful in promoting change in even the most resistant or difficult families. The therapist and observers meet briefly after each session to discuss their observations and interventions. Cases are also discussed weekly in a longer meeting.

Treatment has six stages:

An important paradoxical intervention is to prescribe the symptom. The family is asked to engage deliberately in the symptomatic behaviors. The goal is for the family to rebel against the directive and, in the process, reduce symptomatic behaviors or bring seemingly automatic behaviors under voluntary control as the family engages in the behaviors by choice. This therapeutic double bind promotes progress regardless of how the family responds.

Paradoxical instructions are also used in more general ways. For example, despite the emphasis on the brevity of treatment, families are advised to change slowly or refrain from changing at all. When change is reported, the therapist might advise slowing down, which usually produces more rapid results. Conversely, when rapid change is produced, the therapist might suggest a relapse to the old behavior. Refraining from change often increases the family’s sense of control over behavior.

If the family expresses apprehension about termination, it is done without the formal ending. Any of the unused ten sessions are “deposited” for later use at the family’s request. In practice, most families do not use them.

Treatment Evaluation. The Brief Therapy group stresses the evaluation of treatment outcomes. A group member not involved in the treatment compares the treatment goals to the observable results by determining: if the specific treatment goal was met, what the current status of the presenting complaint is, if the family sought additional therapy, if improvements occurred in other areas of the patient’s life, and if new problems have arisen (in order to address the possibility of symptom substitution).

The group reports that 40% of treatments succeed, 32% show significant improvement, and 28% fail. In some cases, the team did not formulate a goal concretely or specifically enough to evaluate its outcome adequately. In other cases, the changes did not provide relief. According to the authors, these results compare favorably with those of longer-term therapies.

The MRI group is considered the “engineers” of the family therapists, and some critics have described Brief Therapy techniques as manipulative. The Brief Therapy group counters that some influence is necessary to change behavior and that therapists are specialists in influence. By engaging the therapist, the family is saying that change is desired. In their view, it is the therapist’s responsibility to apply his/her skills, drawing on a range of possible interventions, to help bring about that change.

♦♦♦♦♦

Weber, T., McKeever, J., & McDaniel, S. (1985). A beginner’s guide to the problem-oriented first family interview. Family Process, 24(3), 357–364.

Family therapy supervisors Weber, McKeever, and McDaniel, provide a framework with suggested time distribution for guiding therapists through the initial sessions, which serves both as a teaching device and an assessment tool for supervisors to use with beginning therapists. For the authors, the beginning of treatment is critical. In the first interview, the therapist joins with the family by noting the particular organizational style of the family members and selects a therapeutic style that helps family members feel supported and safe. The session structure should provide family members with a sense that the therapist has a plan and will take the lead. The session allows the family’s process to surface and become defined. A therapy contract is developed with emphasis on the family’s goals and desired changes.

If the first contact is by telephone, the therapist: gathers basic information (names, addresses, telephone numbers); asks for a brief description of the problem; identifies members of the family and others who may be involved with the problem; contracts for the first session, including who will be present, date and time, location, and directions to the office or facility; and explains the fees. The individual making the call may be the most highly motivated member of the family. The therapist might wonder why he/she is calling now and how other members of the family would present the reason for seeking therapy.

If the family is not self-referred, the therapist obtains the referring person’s perspective and relationship to the family and clarifies the circumstances of the referral. He/she defines when and how communication and possible ancillary involvement will ensue (obtains releases of information).

Building a Strategy/Making Hypotheses. Tentative hypotheses, generated from the initial telephone contact, form a working framework through which a therapist begins to make sense of the family’s organizational structure, especially as it relates to the presenting problem. These working hypotheses help the therapist develop both a strategy and specific questions for the first interview. However, the therapist must treat the early hypotheses tentatively as he/she gathers additional data and formulates new hypotheses. The early hypotheses are tested and reformulated. The therapist ascertains the family’s place in the life cycle and predicts the tasks and issues surrounding their current phase.

The Interview

Any suggestions offered during this phase should be directed toward the therapy goal and/or directly related to the behavior that was observed in the session. For example, if a younger child continually interrupts a parent, the therapist may suggest that an older child help with the younger one so the parent can focus on the session. Compliment family members when their behavior is constructive, for example, “Tom, you are really helpful with little Jimmy. Thank you.”

The authors provide the following checklist to evaluate the initial interview. The therapist should have done the following:

Following the initial session and the checklist review, the therapist should refine the hypotheses and plan for the next session. The referring person, if applicable, should be contacted. The therapist should decide what information, if any, will be shared. The circumstances of any collaboration should be determined. The therapist should obtain records or other relevant external data.

Haley and Madanes

Jay Haley left the MRI group in 1967 and worked for the next 10 years with Salvador Minuchin (see Chapter 5) and Braulio Montalvo at the Philadelphia Child Guidance Clinic (Nichols & Schwartz, 1998). He then formed the Family Therapy Institute in Washington, DC, with Cloe Madanes. Although Haley’s model is often grouped with strategic models, his work is also clearly influenced by Structural Family Therapy. Like Minuchin and other structuralists, Haley believed that treatment must address not only the symptoms or the presenting problem but also the underlying family structure that maintains the symptoms. Haley’s work is also clearly influenced by Erickson, particularly in his use of directives (between-session tasks) and paradoxical interventions.

Theory of Normal Development and Dysfunction

The Haley–Madanes model is more theoretical than the non-normative MRI Model. Like structural theorists, Haley and Madanes “consider family life cycles … and general systems concepts (e.g. homeostasis, positive feedback) in their conceptualizations of family functioning” (Piercy et al., 1996, p. 51). They contend that symptoms stem from a faulty organization within the family and serve a function in maintaining both family structure and homeostasis. In their view, the hierarchical arrangement of family members is critical. “Haley (1976) suggests that, ‘an individual is more disturbed in direct proportion to the number of malfunctioning hierarchies in which he is embedded’” (cited in Nichols & Schwartz, 1998, p. 360). Madanes adds that symptoms may also function in what she calls incongruous hierarchies, which are “created when children use symptoms to try to change their parents” (Nichols & Schwartz, 1998, p. 361).

Assessment and Treatment

Like MRI brief family therapists, Haley and Madanes are interested in present behaviors and sequences of interactions, using strategic interventions to alter the interactions. They differ from the purely strategic models in that the goals of therapy extend beyond altering the sequences of interactions to include changing the underlying family structure (Piercy et al., 1996).

A prominent feature of the early Haley (1976) model is the strong recommendation that therapists actively plan the therapy from the beginning. The first session is critical. “If therapy is to end properly, it must begin properly” (Haley, 1976, p. 9). During the first session, the therapist and family must define a solvable problem, while the therapist seeks to discover the “social situation that makes the problem necessary” (p. 9). For example, a child’s problem or behavior may function as a reflection of an underlying marital problem.

Haley strongly advises therapists to require all people living in the household or otherwise integrally involved in the problem to be present at the first session. At the same time, he allows flexibility regarding the place of therapy (school, home, and office), the length of the first session, or the fee charged. Given the importance of the first session, Haley developed a detailed, four-stage process and outlined the goals of each stage:

Goals: help family members feel comfortable and relaxed, begin observations of interactions, and make tentative hypotheses about family structure, e.g., who tries to enlist the therapist to his/her side? How do the parents discipline the children? Hypotheses should be tentative since early observations may reflect a tendency for the family to act in the way they think the therapist expects.

Goals: continue to observe and make mental hypotheses about hierarchical structure and triangles but does not “interpret” family interactions to them. Notice disagreements in the explanation of the problem, which will provide the basis for the interactive discussion to follow. Take charge of the session by, for example, bringing in under-involved members, preventing an overly talkative member from dominating, sitting near a child reluctant to speak, or moving a child closer to the parent to whom he/she is the least close. These interventions are strategic since the family is prevented from repeating their previous patterns of interactions.

Goals: test hypotheses; observe sequences and the structure governing behaviors such as malfunctioning hierarchies, coalitions, quality of parental functioning, and the like.

Goals: to delineate a solvable problem that can be addressed in therapy. If articulated in such a way, the therapist and the family know when treatment is completed.

Haley provides an extensive checklist to evaluate the first session.

Directives. In his early work, Haley designed tasks, called directives, for three purposes:

Directives should be concise and may involve part or all of the family. The task may begin during the session with instructions to continue at home. As in all strategic therapy models, the therapist tracks the sequences involved in the problem and assigns directives that alter the sequence. However, for Haley, the goal is to alter the family structure in order to correct malfunctioning hierarchies, often by strengthening the parental unit. The family’s attempts at directives should be reviewed in the next session. For example, if a mother is behaving in a helpless way with her children, forcing the exasperated father to assume a primary parenting role, the father may be directed to instruct the mother nightly on parenting techniques. The mother will rebel against being instructed and thus become a more effective, and equal, parenting partner.

Haley’s earlier therapy was also characterized by the use of ordeals (Nichols & Schwartz, 1998). Ordeals are directives that are aimed at making symptoms harder to maintain than to relinquish. An ordeal requires clients to do something they do not want to do but that would nevertheless benefit them in some way. (For example, a husband might be instructed to give a present to a brother-in-law with whom he has a strained relationship each time the symptom occurs. If the husband finds the task unpleasant enough, the symptom may disappear, and the relationship might improve as a result.) Like all directives, ordeals are aimed not only at symptom relief but also at family restructuring.

In contrast, the most recent Haley–Madanes therapy model, strategic humanism, is “more oriented toward increasing family members’ ability to soothe and love than to gain control over one another” (Nichols & Schwartz, 1998, p. 374). Madanes has separately elaborated her own theories and strategies.

♦♦♦♦♦

Madanes, C. (1980). Protection, paradox and pretending. Family Therapy Process, 19(1), 73–85.

While still considered a strategic therapist in view of her approach to solving problems within families, Madanes proposes a structural explanation for psychopathology in children. In a well-functioning family, parents are hierarchically superior to their children and can usually “pull themselves together” to help their child, even despite an existing marital conflict or one parent struggling with an issue.

In dysfunctional families, the child has become hierarchically superior to one or both parents—a powerful but detrimental position. He/she becomes the parents’ benevolent protector by taking on symptoms, causing the parents to focus on the child’s behavior rather than their own problems. The child and one parent may form a coalition against the other parent or a grandparent—a structural misalignment referred to as triangulation. The child’s problem behavior provides a bond that holds the parents together. Conflict between the parents, or even family issues inherited from prior generations, can be expressed through the child. Regardless of how the parents respond to the child (e.g., soothing, demanding, punitive, concerned), it shifts the parents’ focus on their child and away from their own issues.

Assessment and Treatment

The first task of treatment is to determine who is being protected by the child’s behavior and how. The therapist then prescribes a directive designed to change the pattern of interaction to reestablish the parents in a superior position by helping the parents take back power from the child. The changed structure no longer supports the child’s problem behaviors. The emphasis is not on helping the family understand how or why the problem behavior is occurring but rather on solving the problem. Directives are developed to fit the unique needs of the family. The strategic paradoxical interventions are: dramatizations, pretending, and make-believe play.

Dramatizations. A parent is directed to request that his/her child intentionally perform the problem behavior. In this intervention, the relationship between the parent and the child is based on benevolent helplessness in which the child’s symptom helps the parent by diverting attention from the problem as the parent helps the child overcome the symptom. For example, a mother is worried she may lose her job (the real problem) and the child develops a headache, as an attempt to protect the mother and “solve” her problem. The strategy works in that the mother’s attention shifts from her own problem to her son’s headache. To alter the pattern, Madanes directs the parent to encourage the child to have the symptom. In this way, the symptom will not draw as much parental attention, no longer serves a purpose, and will usually be dropped. Subsequently, the mother’s fear will resurface, and she can address the real problem with the therapist’s help.

Pretending. Madanes directs parents to ask the child to pretend to have the symptom, while the parents pretend to help the child. This intervention turns the child’s need to act out and the parent’s need to help into a kind of game. For example, a child developed stomachaches in order to receive affection from his grandmother. The symptom not only provided a way for the grandmother and grandson to express mutual tenderness, it also added an element of drama to the grandmother’s otherwise routine existence—but at a cost to the child. During therapy, the child was asked to pretend that he had a stomachache and the grandmother to pretend to care for him. They were tasked with doing this at home every day for one week. Through the directive, the child and grandmother could still be affectionate, the grandmother continued to be needed and loved, but since the stomachaches were no longer necessary, they disappeared.

Make-Believe Play. When a child protects his/her parents through symptomatic behavior, he/she is helping them covertly. Instead, Madanes asks parents to engage in a make-believe play pretending that they need the child’s help and the child to make-believe help them. Since the parents explicitly ask for help and the child overtly helps them, there is no need for the covert symptomatic behavior. Additionally, when parents intentionally assume an inferior position, they may feel at odds with what is appropriate and reassert their superior position.

For example, a mother with several children was in fear of losing her welfare benefits because of a live-in boyfriend. Her son sensed her fear and helplessness and developed night terrors. His nightmares were both a metaphor for the mother’s fears and an attempt to help her since she had to set aside her own fears to comfort him.

The family was asked to make believe that the mother was afraid because she thought someone was breaking into the house and that she needed the child’s help. The therapist designed a make-believe play in which the son was to protect his mother. They were directed to perform this play every night at home. If the mother heard her son screaming in the night, she was to wake him and his sisters up and replay the drama, regardless of the time. This intervention was designed to encourage the son and his mother to change the way they were protecting each other by transforming the mother’s need for help and the son’s helpfulness into play. Follow-up sessions found the night terrors had ceased, the mother was in a productive job and working on her relationships, and the son was involved in sports and school. The goals of therapy were reached in two stages. At first, the paradoxical directive blocked the ways in which the son’s symptoms served to help his mother avoid confronting her problems. Later, the mother regained her appropriate position in the family hierarchy.

Most recently, Madanes has moved away from strategic and structural models, closer to the work of Virginia Satir (see Chapter 6). She now hypothesizes that family problems result from a “dilemma between love and violence” (Madanes, 1990, cited in Piercy et al., 1996, p. 52). In this model, family members have four aims, and problems in each area tend to lead to different types of symptoms: (1) to dominate and control, leading to behavioral problems such as delinquency and drug abuse; (2) to be loved, leading to depression, psychosomatic problems, phobias, and eating disorders; (3) to love and protect, leading to suicide, abuse, neglect; and (4) to repent and forgive, leading to incest, murder, and abuse (Nichols & Schwartz, 1998; Piercy et al., 1996). The goal is to intervene in the family through a 16-step program guiding the family toward love and away from violence. She also includes “growth-oriented objectives like balance, harmony and love” (Nichols & Schwartz, 1998, p. 365).

♦♦♦♦♦

Hoffman, L. (1981). “Chapter 11: Breaking the symptomatic cycle,” in Foundations of family therapy (pp. 197–218). Basic Books.

Hoffman presents three family cases, each illustrating symptomatic sequences that influence the schismogenic tendency of relationships. The families are each seen by therapists with different theoretical perspectives; but, in each case, the treating therapist interrupts the dysfunctional recursive sequence to help the family move to a desired level of functioning. Changing the symptomatic sequences involves two types of corrective action: first- and second-order changes.

Behavioral sequences in families tend to fluctuate within a limited acceptable range. Therapeutic modifications that occur within this range are called first-order changes. Such limited adjustments are often sufficient to solve many problems. For example, a mother striving to improve her family’s nutrition may broaden her choice of what to serve for dinner so long as it does not exceed her family’s preferences.

Second-order change falls outside the accepted range of behaviors and is often precipitated by new circumstances and/or the family’s natural evolution through developmental stages. Second-order change is often preceded by a major shift in family rules and may result in a fundamental reorganization of family structure.

When first-order changes no longer bring about adequate solutions, many families are able to make a second-order change. But in other circumstances the attempted solution may itself become a problem. Thus, when these families seek treatment, the therapist might either provide direct advice or help the family generate alternate behaviors, which may fall within the realm of first-order change. However, when the cycle that maintains the problem is too rigid or a structural change is called for, the therapist may need to consider second-order change. For example, a parent who is struggling with an adolescent child’s eating habits may try to force him/her to eat particular types of food. The child responds by refusing to eat, reinforcing the parent’s concern that the child is not eating properly, compelling the parent to take even greater control. The parent’s escalating attempts to control the child and his/her increasingly stubborn refusal have become the problem. Watzlawick, Weakland, and Fisch (as cited in Hoffman, 1981) refer to this recursive cycle as the game without end.

The therapist might first try “a little push,” for example, advising the mother to allow her son greater freedom to determine his own diet. The push may not work, and the therapist may need to help the family focus on what is at the root of the symptomatic cycle, for example, explaining that since this child is now an adolescent, he has less need for guidance regarding food. Further, the parent may need to allow the adolescent to take on more responsibility in this and other areas, and the parent may need to loosen control. This realization can disrupt the cycle, resulting in second-order change, new rules, and a new organization of the family.

Families become stuck in repetitive patterns of behavior, however, and the therapist may need to actively interrupt the cycle to enable the family to make a change. He/she must first identify the cycle. Some cycles are obvious, while others are not easily recognized, such as with psychosomatic illnesses or communication disorders associated with psychosis. In those situations, the therapist can gather information on how the condition or problem is managed, which often reveals the cycle. In most cases, these problems and recursive sequences are a reflection of a dysfunctional triangle which brings the child into what is actually a mirror-image disagreement between the parents. In such cases, the solution to the problem often involves helping the parents get together so they can make the child behave.

The three cases Hoffman discusses are briefly outlined below:

Minuchin first arranged to meet with the parents and anorectic daughter over lunch, then excused himself to observe them. The parents consistently took opposite positions.

One reasoned with their daughter, while the other demanded she eat. When the reasoning parent started to become more demanding, the demanding parent became gentler and began to reason. The configuration that was revealed was one of a disciplinary father, a permissive mother, and a daughter caught in the ongoing mirror-image disagreement. Each person covertly gives cues for behaviors to one another. The family members could not stop the spiraling cycle on their own, and the cycle kept them from launching their daughter into the adolescent departure stage.

Minuchin’s goal was to provoke a runaway positive feedback loop, which throws the family out of equilibrium and increases stress. He removed one member of the triangle at a time, leaving each parent to try to deal with the daughter on his/her own. This eliminated the constraints the parents placed on each other and forced the behaviors in the sequence to exceed their usual limits. The mother became feebler and the father more forceful and violent.

Minuchin then reframed the problem as one in which their daughter was stronger than them. Thus, he moved the parents from a detouring-benevolent triangle in which the daughter was conceived of as “sick” to a detouring-attacking triangle in which she is perceived as “bad.” The parents were then able to talk to Minuchin about the more threatening problems in their relationship, freeing the daughter from the cycle. The girl quickly resumed normal eating habits. The family had undergone a second-order change.

Hoffman reports on a family in which the parents were locked in a battle with their son over his smoking. Fulweiler was the therapist, with Haley acting as the supervisor. Again, the family members formed a triangle: a domineering father, a rebellious son, and an ineffectual mother who tended to side with her son. The father would attempt to force his son to quit smoking, the son would rebel, and the mother’s lack of support for the father undermined his position and caused him to retreat. As in Minuchin’s case (discussed above), the sequence diverted attention away from the marital problems and placed the son in the middle of a mirror-image struggle between the parents. If he quit smoking, he would be supporting his father against his mother; if he continued to smoke in defiance of his father, he would be supporting his mother.

Fulweiler’s strategy was to use blocking maneuvers to disrupt the sequence. He used a technique in which he exited and re-entered the therapy session without warning, each re-entry serving a strategic purpose. At first entrance, he helped the mother clarify her position; at the second entrance, he highlighted a marital disagreement; the third served to bolster the father’s authority, and the subsequent entrances to stop the father’s tendency to adopt the role of a victim. According to Haley, this family illustrates standard roles: over-involved parent, peripheral parent, with the child functioning as a secret agent defying the over-involved parent on behalf of the peripheral parent.

Fisch argues that even a very small change in a system can produce a wide-reaching effect. He presents a case in which parents were struggling to manage their children, one of whom they characterized as willful and obstinate and the other as well behaved. The parents also characterized their marriage as loving and close. In the family’s interactional pattern, the mother got into daily battles with the daughter and the father intervened to subdue the daughter. Fisch told the parents they were having trouble controlling their daughter because their responses were too predictable. To interrupt the dysfunctional sequence, he instructed the father to give his daughter a penny during the next mother-daughter battle. Following the intervention, a previously hidden, covert mirror-image disagreement emerged when the mother expressed little hope for the daughter she perceived as impossible, while the father held higher expectations. In this type of intervention, the therapist creates confusion aimed at breaking the cycle and uncovering mirror-image disagreements that have been hidden.

In sum, Hoffman illustrates several ways to encourage first-order change, where possible, by giving the family a little push, offering advice, or suggesting alternative behaviors. Where second-order change is necessary, therapists might escalate positive feedback loops, increase stress, create therapeutic confusion, or interrupt and alter rigid recursive cycles.

♦♦♦♦♦

Hoffman, L. (1980). “Chapter 3: The family life cycle and discontinuous change,” in E. Carter & M. McGoldrick (Eds.), The family life cycle: A framework for family therapy, pp. 53–57. Gardner.

In this article, Lynn Hoffman integrates common, though sometimes ignored, observations about the surprising ways families change, together with the scientific research on change process models, particularly those drawn from biology or physics. She outlines mechanisms for change that expand the family system’s cybernetic view, reexamines the meaning of symptoms, and suggests ways in which therapists can intervene with families in crisis to foster creative leaps in functioning. Finally, she relates these observations to the family life cycle.

The systems model of families is, at its core, a homeostatic model. Most behaviors, particularly symptomatic ones, are understood as serving to keep the family functioning within a relatively stable range with respect to characteristics such as closeness, independence, power structures, and the like. When the system threatens to exceed that range, feedback mechanisms work to return the behaviors into a familiar static state. Accordingly, the model predicts that when change occurs, it happens slowly. Hoffman suggests that the model is compelling, in part, because it helps to explain the apparent “stuckness” that family therapists frequently observe. Family members’ tenacious resistance to change implies that the symptom is needed to maintain equilibrium.

But, as Hoffman observes, families often do not change in smooth, continuous progression. Instead, they make sudden, often creative, shifts—called discontinuous changes—either on their own or in therapy. Platt (cited in Hoffman, 1980) distinguishes three kinds of change, depending on the type of system. If a system is externally designed, like an engine, change must be introduced by someone outside the system, like a mechanic. If it is internally designed, like a flower that contains a genetic blueprint, change occurs through mutations of the genetic material. In human systems that follow a self-maintaining design, change can occur in the form of a transformation—a shift or change in how the parts are organized within the system. Hoffman draws on the writings of Paul Dell and Harry Goolishian, Bateson, and Ashby to explain sudden, discontinuous change.

Discontinuous Change Mechanisms

Dell and Goolishian’s work is grounded in a notion of change that they refer to as “order through fluctuation” (as cited in Hoffman, 1980, p. 53). According to this view, families maintain homeostasis so long as the pressures on the system are relatively stable. However, events often put considerable stress on the family, pushing the family to function beyond its previous level of functioning, risking breakdown unless it reorganizes. The growth process is facilitated by a feedback mechanism described by physicist Ilya Prigogine, called “evolutionary feedback, … [which is] a non-equilibrium ordering principle” (as cited in Hoffman, 1980, p. 53, emphasis added) that operates when fluctuations exceed the system’s homeostatic range. Systems tend to operate inside of a range of stability within which fluctuations are damped down, and the system remains largely unchanged. Should a fluctuation become amplified, however, it may exceed the existing range of stability and propel the entire system into a new dynamic range of functioning (pp. 53–54).

Dell and Goolishian contend that families cannot be fully understood using a “cybernetic analogy on a mechanical model of closed system feedback” (as cited in Hoffman, 1980, p. 51). Instead, they claim that the cybernetics of living systems is different, as vividly illustrated by their discontinuous changes and sudden leaps into new levels of integration.

Drawing on Bateson’s ideas, Hoffman notes that he also emphasized the human capacity to move beyond simple behavioral replication to creativity, art, learning, and change. The “processes of change feed on the random. The essence of learning and evolution is exploration and change” (p. 53).

According to Ashby, families seem to operate on a bi-modal feedback mechanism. A system remains unchanged so long as its internal or external environment is stable; but when a fluctuation exceeds the range of stability, the system must respond in some new way. The system either breaks down or makes a leap into new levels of functioning. The change results in a new set of patterns that, like the old one, is also bound by rules, and it, too, remains unchanged so long as the environment is stable.

Hoffman summarizes the process as follows. In response to environmental changes for which the system is not yet designed, patterns of responding that previously served the family well begin to fail. The family tries new solutions, many of which are necessarily abandoned, leaving the system in a state of confusion. As its homeostatic tendencies yield increasingly wild corrective attempts, the system enters a period of crisis. Finally, the system either breaks down or “may spontaneously make a leap to an integration that will deal better with the changed field” (p. 56).

Such discontinuous changes, like symptom development, often occur during stressful times. Changes in family composition are particularly demanding. Hoffman distinguishes crises of accession, when someone joins the family (marriage, birth), and crises of dismemberment, when members leave (divorce, death).

Symptoms and rapid changes in family functioning also tend to emerge during the negotiation of developmental stages. Pressures arise, for example, as an adolescent reaches a new, more independent maturational level. While there is no uniform agreement on the number of developmental stages, Hoffman identifies the major categories as “courtship, marriage, advent of young children, adolescence, leaving of the children, readjustment of the couple, and growing old and facing death” (p. 58).

Hoffman describes the mechanism for discontinuous change that draws from Ashby’s work on similar changes in the physical world. Of the several types of change mechanisms Ashby reports, the most salient to understanding change in families is “step-function,” which consists of “intervals of constancy separated by discontinuous jumps, like a set of stairs” (as cited in Hoffman, 1980, p. 58). These discontinuous jumps occur suddenly as the system exceeds its capacity, analogous to a fuse blowing when the load of a circuit surpasses its limit. Without a fuse, the whole system would break down, but once the system is back within its limits and the fuse replaced, the system is functional again. He refers to this as a step mechanism.

Applied to families, the stressor may be temporary. For example, a teenage daughter runs away from home when she fails a mid-term exam. When the family discovers she is safe at her friend’s house and quickly hires a tutor, the ‘circuit load’ returns to tolerable levels, and the ‘fuse’ is replaced—the stress on the system is reduced and the family can resume functioning. In other cases, however, the stressor may be permanent and require either rapid creative changes or result in a breakdown. For example, the ‘fuse’ may blow when Dad loses his job. The family might resort to old ways of functioning—replacing the fuse, so to speak, and fixing the circuitry. If Dad becomes depressed and starts drinking, the system may break down as former patterns of behavior fail, and the fuse blows again. Hoffman describes such a breakdown as a “step mechanism signaling the failure of the family’s homeostatic mechanisms” (p. 60). Alternatively, Dad may decide to take an early retirement and expand his gardening hobby into a business, allowing the family to reorganize itself around the change.

Thus, in this model, “symptomatic displays could … be thought of negatively as aborted transformations—a failed leap—or positively as negotiations around the possibility of change [or even as a] compromise between pressures for and against change” (p. 61). Symptoms can function to prevent change from occurring too rapidly, albeit often at the expense of one family member. They may also help maintain pressure on the family to make needed changes. But, as Hoffman points out, while a shift to a symptomatic pattern may serve as an immediate solution, it neither constitutes nor reliably leads to a more functional transformation.

Implications for Treatment

Hoffman explores the implications for these ideas in helping families negotiate environmental stressors and developmental stages so that they make the necessary creative transformations and prevent the symptom from spiraling into chronic behavior patterns. That is, how does a therapist prevent morphostasis and encourage morphogenesis? It is important that therapists not interpret conflict or apparent dysfunction as an omen of impending disaster, but rather “that pressure toward a new and more complex integration is mounting” (p. 55). Understanding various aspects of communication and the effects of messages is also important. Hoffman discusses the related concepts of paradoxical injunctions and double and simple bind communication.

A paradoxical injunction is a statement “that intrinsically contradicts itself unless teased apart into a ‘report‘ level and a ‘how this report is meant’ level, the second level inclusive of the first” (p. 62). This concept corresponds to what the Palo Alto researchers called the double bind, in which a message is given to another person requiring contradictory, mutually exclusive responses (i.e., do something and do not do something). This concept was initially used to describe the communication patterns in families in which one member was diagnosed with schizophrenia. Hoffman notes, however, that such contradictory messages are common throughout society and do not usually result in madness. Instead, she refers to them as sweat boxes since they indicate a mild or severe threat to the continuity of the relationship and the system. She asserts that these types of pressures may be a necessary precondition to morphogenesis.

However, Hoffman makes an important distinction between a double bind message and a simple bind message. In a double bind, whatever response the message recipient gives, his/her “leap in behavior” is disconfirmed; by contrast, in a simple bind, a new response is rewarded. “In other words, the double bind is a simple bind that is continually imposed, and then continually lifted” (p. 64).

Hoffman illustrates the distinction with the example of a teenager whose mother exhorts him to behave in a more mature way. If he disobeys and stays immature, he fails. If he acts more maturely, however, he is being an obedient child in a mother–child relationship, and thus fails again. Since neither response is rewarded, the child is caught in a double bind. If, however, mother and son spontaneously find themselves relating to one another more as adults, the son’s new behaviors are rewarded. In this case, the injunction is a simple bind, and the two transform their relationship with a creative leap.

Because stress, development, and change cannot be avoided, Hoffman argues that family therapy must make a radical departure from models that support homeostasis. Rather than attempting to avert crises, the therapist should expect, even welcome, crises as opportunities to foster creative transformations.

The knowledgeable clinician … will know that these behaviors are expectable concomitants of family change. He or she will seek to disrupt the homeostatic sequence that forms about a symptom so that pressure for change will be allowed to build and a transformation will hopefully take place that makes the presence of a symptom unnecessary (p. 67, emphasis added).

Neuro-Linguistic Programming (NLP)

This model has its roots in the works of Gregory Bateson, Milton Erickson, and Virginia Satir. It evolved from Richard Bandler and John Grinder’s extensive study of audiotapes and movies of Satir and Erickson’s clinical work. NLP examines the relationship between language and reality, following the ideas of Alfred Korzybski, “A map is not the territory it represents, but, if correct, it has similar structure to the territory, which accounts for its usefulness” (Korzybski, 1958, p. 58). The NLP model attends to how language reflects a person’s ability to change or not to change. Through the structure of language, concepts such as generalizations, deletions, and constraints emerge that shape belief systems and life choices. The NLP therapist carefully assesses the structures found in a person’s language of experience. Then he/she initiates therapeutic conversations and techniques, similar to trance or hypnotic induction, in order to restructure the client or family’s beliefs, instill a sense of possibilities, and increase the likelihood of change. The Structure of Magic I (1975) and II (1976) and Frogs into Princes (1979) by Bandler and Grinder are three volumes that detail the methods and techniques that NLP practitioners use.

Milan Systemic Family Therapy

Overview

Systemic therapies originated in Milan, Italy, with Mara Selvini Palazzoli, Luigi Boscolo, Gianfranco Cecchin, and Giuliana Prata. Like the strategic models of MRI Brief Therapy and the Haley–Madanes approach, the Milan Systemic model grew out of Bateson’s work on cybernetics, in which problems were viewed as being maintained by interactional sequences. While Erickson further influenced MRI and Haley–Madanes, as noted above, the systemic therapies, at least originally, held more strictly to the Bateson formulations (Piercy et al., 1986). Their 1980 seminal article is summarized below.

The original Milan Group, started by Mara Selvini Palazzoli, initially treated severely disturbed children using traditional psychoanalytic methods. Increasingly frustrated by their lack of progress in working with children alone, the group became increasingly influenced by the writings of family therapists who worked with whole families instead. They first tried extending the psychoanalytic model to families. After reading Bateson’s work, Boscolo, Cecchin, and Selvini Palazzoli broke from the earlier Milan Group configuration and formed The Center for the Study of the Family with a goal of working in the new systems model. Watzlawick was originally a consultant to the group (Nichols & Schwartz, 1998). In the early years, the team conceptualized family problems as being maintained by homeostasis—a tendency to resist change—and devised paradoxical interventions to counter this tendency.

Theory of Normal Development and Dysfunction

Like the MRI team, the Milan Group adhered to a “non-normative” stance. Further, they maintained a neutral stance toward therapy outcome, trusting that if families were helped to see new ways of understanding their problems, they would find better ways of organizing themselves, without a need for reference to norms. In this perspective, symptoms simply function to preserve family homeostasis and are being maintained by recurring interactional sequences.

The Original Milan Model

The first model was strongly influenced by the MRI Strategic methods. Families were treated by a male–female dyad of co-therapists and observed by other team members. Each session had five parts:

The sessions were held one month apart to give families time to react to the interventions, and the total number of sessions was usually limited to ten. Two basic interventions, positive connotation and rituals, characterized the early model.

Positive Connotation. Positive connotation is the hallmark of the early Milan Systemic Model. The Milan Group believed that people could not easily change under the influence of negative connotation. For example, diagnostic labeling (a negative connotation) implies causality and implicates the person in the diagnosis. Positive connotation, by contrast, avoids linear causality and blame by assigning a positive motive or value to each family member’s behavior. Their initial intervention technique was similar to reframing (used by the MRI therapists), since the symptom was assumed to serve a protective function, and the goal of the intervention was to alter the way the symptom was viewed by the family. However, these therapists objected to the technique of reframing to the extent that family members feel blamed for creating problems in their families. Positive connotation “eliminated the implication inherent in such reframes that some family members wanted or benefited from the patient’s symptoms” (Nichols & Schwartz, 1998, p. 375), which might lead to greater resistance.

Positive connotations should be distinguished from reframing. Reframing can be positive or negative. It is directed toward one family member and ascribes meaning to a behavior. A positive connotation, on the other hand, always addresses every family member’s part in the circular process that maintains the problematic interactions.

By contrast to the strategic therapies, in the original Milan model, the problem is not viewed as “useful” so much as something the family has gotten used to. In a positive connotation, the family might be told, for example, that the patient should

continue to sacrifice himself by remaining depressed as a way to reassure the family that he will not become … abusive … like his grandfather. Mother should also remain over involved with [her son] as a way to make him feel valued while he sacrifices himself. Father should continue to criticize [mother and son’s] relationship so that mother will not be tempted to abandon [her son] and become a wife to her husband (Nichols & Schwartz, 1998, p. 375).

The Milan Systemic model originated as a meta-consultation model. Treatment often included all people who might be part of maintaining the problem. For example, if the family had been referred by another therapist who, in the view of the Milan team, had become part of the impasse preventing change, the therapist might be brought in with the family (who might also be invested in keeping the therapist enlisted in its coalitions and the maintenance of its homeostasis) and would be included in the positive connotation. For example, the therapist might be thanked for helping the family by failing to require changes. Later in the development of the model, the Milan therapist dyad might also be included in a positive connotation by the observing team.

Rituals. Rituals are interventions that enhance a positive connotation or require the family to either exaggerate or violate family rules. For example, to exaggerate a positive connotation, a family might be asked to thank the symptomatic family member for having the problem. A family that maintains loyalty to an extended family to its own detriment might be asked to violate the family rule by holding secret meetings.

Over time, the team increasingly turned away from paradoxical interventions and focused instead on the processes that occurred during the therapy session itself. They developed interventions based on the concepts of hypothesizing, circularity, and neutrality (Piercy et al., 1986). In 1979, the Milan team split. Selvini Palazzoli and Prata formed one group and Boscolo and Cecchin another. Selvini Palazzoli and Prata focused on interrupting destructive family games (Nichols & Schwartz, 1998), leading to the development of the invariant prescription (see below) for treating severe pathology. By contrast, Boscolo and Cecchin stayed with the concepts of hypothesizing, circularity, and neutrality (Piercy et al., 1986) and became interested in changing family belief systems and underlying epistemologies, which eventually led the way into the Solution-Focused and Narrative therapies (see Chapter 7).

Selvini Palazzoli and Prata

This team hypothesized that power games in the family lead to the development of symptoms intended to protect the family. According to their theory, psychotic games develop in families in six stages (Piercy et al., 1996):

Assessment and Treatment

Before the 1990s, when Selvini Palazzoli and Prata were engaged in systemic work, their goal was to help parents form a stable and thereby alter patterns of interactions among family members. The intervention, the invariant prescription, was the same for all families (Nichols & Schwartz, 1998). The team directed the parents in the formation of a secret coalition. First, the parents met with the therapists without the knowledge of other family members and then began taking progressively longer secret trips, eventually spending several days away without telling other family members. They were asked to keep notes on family members’ reactions to review with the therapists.

By the mid-1990s, although still influenced by her years of systemic work, Selvini Palazzoli returned to long-term psychodynamic treatment models for individuals and families. “This new therapy revolves around understanding the denial of family secrets and suffering over generations” (Nichols & Schwartz, 1998, p. 376).

Boscolo and Cecchin

Boscolo and Cecchin became interested in the processes that occurred during therapy sessions. They believe that when family members gain new information in a session, providing them with an understanding of the tacit beliefs and rules under which they operate—or their epistemology—they are stimulated to find a new epistemology that allows new ways of operating. The goal of therapy is simply to introduce new information rather than set specific goals for changes (Piercy et al., 1996). The therapist asks the family questions that “are designed to decenter clients by orienting them toward seeing themselves in a relational context and also seeing that context from the perspectives of other family members” (Nichols & Schwartz, 1998, p. 376). The therapist is curious about how the family system operates, but indifferent to any particular outcome because to favor one outcome would unduly pressure the family (Nichols & Schwartz, 1998). Instead, the therapist generates multiple new hypotheses to help the family find different ways of viewing and understanding their problems.

The model is characterized by the concepts of hypothesizing, circularity, and neutrality, which originated in the Milan Group’s work prior to the split.

Assessment and Treatment

Hypothesizing is an assessment tool through which the therapist begins an exploration into the family system and invites the family to join him/her in the investigation. Hypotheses must be systemic; that is, they must take into account all relational components of the family. The working hypothesis guides the therapist’s circular questioning. “Without [a] hypothesis [the therapist’s] questions will lack a coherent meaning and bring no new information to the family” (Piercy et al., 1996, p. 61).

Alternate hypotheses develop through the questions the therapist poses to the family. In turn, responses from the family generate new hypotheses by the therapist, which leads to new questions, more responses, and new hypotheses. In this recursive process, all hypotheses are considered equally valid (Piercy et al., 1986) so long as they provide new information about how the family system operates.

Influenced by Bateson’s work, Boscolo and Cecchin believe that premises, values, or guiding principles underlying family behavior might be unconscious. In forming hypotheses and questions, they look for a premise or myth that organizes and sustains problem-related behaviors. If the premise can be shifted, change might occur alongside corresponding changes in beliefs.

Circularity refers both to the nature of member-to-member interactions and to the form of interactions between the therapist and the family. Any individual family member’s behavior must be understood as part of a circular sequence of behaviors but not its origin (as would be assumed under a model of linear causality).

Circular Questioning is the primary therapy interview technique through which circularity is explored. Most interactions between the therapist and the family take the form of questions and responses. These questions are guided by the therapist’s hypothesis and invite responses consisting of relational descriptions of family interactions. This helps family members see the perspectives of other members. For example, a father may be asked how his wife sees her relationship with their son, or a child might be asked what might happen to his symptomatic brother if their mother and grandmother didn’t fight so much. Circular questions also explore aspects of family interactions such as the intensity, timing, and frequency of the problem (e.g., Did that occur before or after? How much? How often?).

Neutrality (Curiosity) and Irreverence. Neutrality was the term originally used to describe the therapist’s attitude toward hypotheses generated in treatment. It has since been reframed as curiosity and represents the basic therapeutic stance. The therapist conveys an attitude of curious exploration when asking questions or responding to family members’ answers. Recently, Cecchin suggests that therapists also convey irreverence, that is, they should not be inclined toward any single set of rules or beliefs governing family interactions and should encourage a similarly irreverent attitude in family members (Nichols & Schwartz, 1998). One way the therapist may encourage irreverence and/or a more flexible view of family beliefs, is using the odd-day/even-day ritual, in which the therapist would issue a directive that on odd days one set of opinions would be true, but on even days, the same set would be false. The directive for the seventh day is to act spontaneously.

The therapist is also neutral with respect to his/her relationship with each family member, being careful not to form coalitions or take one side against another. He/she avoids a moral or judgmental position toward a family’s ideas or preferred outcome, following Boscolo and Cecchin’s belief that therapy could and should only perturb or disrupt the system, not direct the family toward any particular outcome.

♦♦♦♦♦

Boscolo, L., Cecchin, G., Hoffman, L. & Penn, P. (1987). “Introduction,” in Milan systemic family therapy: Conversations in theory and practice, pp. 3–28. Basic Books.

This chapter describes the authors’ evolution from psychoanalytic to family systems orientation and the changes that have occurred in their model over time.

Boscolo and Cecchin joined Selvini Palazzoli’s group in Milan, Italy. The group originally treated severely disturbed children and their families using the psychoanalytic model but found this model discouraging and frustrating and the treatment ineffective and prolonged. They turned to the work of the Mental Research Institute (MRI), particularly Bateson’s ideas on double bind communication, elaborated most comprehensively in his 1972 book, Steps to an Ecology of Mind.

The MRI therapists focused on relationships rather than individual pathology. The Milan team expanded on the MRI work guided by their belief that changes in symptoms and problems proceeded from changes in interactional patterns. In their view, pathology derived from double-bind communications with their contradictory messages. Schizophrenia, for example, was seen as a “mistaken epistemology that any one person can unilaterally control relationships” (p. 5). This epistemology implies linear causality and leads to one person trying to exert control, which, in turn, leads to the other trying to re-exert control in an endless and fruitless game. Based on their own work with families with a member diagnosed with schizophrenia, the Milan Group concluded that: (1) the families are involved in unacknowledged family games; (2) through these games, family members try to control each other’s behavior; and (3) the therapist’s job is to discover the games and stop them.

Evolving from the MRI game and coalition theories, the Milan Group, in its early years, assumed an adversarial style—the family against the therapists—and focused on resistance to change. The therapist did not challenge the family directly but rather used “creative deceptions—paradoxical prescriptions—that would bypass … resistance” (p. 7). A family could hardly resist changing if a therapist instructed them to continue the behaviors they were already engaged in, thereby eliminating the very conditions under which resistance emerged.

At the heart of the early Milan treatment model was positive connotation, an outgrowth of the MRI symptom prescription technique. “A positive connotation is a message to the family from the therapist[s] that the problem is logical and meaningful in its context” (p. 4). The Milan Group argued, however, that simply prescribing the symptom “negatively connoted family members’ anti-symptom views” (p. 7) and, insofar as the symptomatic member was exonerated, tended to leave other family members feeling at fault. Rather than focusing solely on the symptom, the Milan team also positively connoted the behavior of all family members, framing each member’s actions as serving a function in the system. The intervention was directed not at an individual but at the self-maintaining tendencies of the system as a whole. Prescribing the symptom in relation to its social context helped reduce resistance, supporting systemic reorganization. The method was in keeping with the emerging “nonlinear, systemic consciousness that was to distinguish the Milan method from previous approaches in the family field” (p. 8–9).

In their early treatment approach in the 1970s, the team was divided into two male–female dyads. One would interview the family while the other observed from behind a one-way mirror. Families were seen weekly for ten sessions. The team later changed to having one member of the team with the family and another observing, and the time between sessions lengthened to monthly. The use of observation teams (O-Teams) originated in the Milan Group and continues to offer a way to bring in trainees and provide ongoing supervision. Rather than being fixed in one method, the teams have shown the ability to experiment and evolve into different forms, e.g., adding an O-team into the mix.

Although teams operating in the Milan Group tradition exist worldwide, they have been less numerous in the United States, where they have often been absorbed into strategic teams working in the MRI Model or using an Ericksonian approach.

During the 1970s, the Milan Group was again influenced by Bateson’s writing, particularly the idea of cybernetic circularity as a model for human systems. Boscolo and Cecchin began a dramatic shift in their work from strategic to a systemic view. Their work and teaching then resulted in the method of circular questioning and the publication of Hypothesizing — circularity — neutrality: Three guidelines of the conductor of the session (1980) by Selvini Palazzoli, Boscolo, Cecchin, and Prata (summarized below).

The team translated Bateson’s ideas of cybernetic circularity into a systemic treatment model. Hypothesizing is the assessment process; circularity and circular questioning is the technique; and neutrality is the basic therapeutic stance. Their systemic hypotheses account for all the elements of the family problem and how they are connected. It is a model grounded in the emphasis that the team and the family are engaged in a research project together. The validity of the hypothesis is less important than its utility in providing new information about how the family operates and helping the family progress.

Circular questioning, based on Bateson’s idea that “knowledge is always apprehended by mechanisms that scan for difference” (p. 11), provides feedback to the therapist regarding family relationships. In this technique, the therapist conducts the investigation and hypothesizing based on information gained about relationships in the family, about differences, and change. The questions fall into several categories: differences in perceptions of relationships (who is closer to grandfather), differences in degree (how bad was the fighting this week), now/then differences (did the fighting start before grandfather died or after), and hypothetical or future questions (if grandfather had not died, how would life be different).

Therapeutic neutrality grew from Bateson’s assertion that all parts of the system should be accorded equal weight. In being neutral, the therapist regards each person in the system as equal and each person’s viewpoint as valid. The therapist does not take a moral position with regard to any part of the family process. The therapist can observe a variety of collusions and coalitions without being inducted into any of them. Neutrality allows the therapist to maintain a hierarchically superior (meta-level) position without being authoritarian, in contrast to the more directive styles of Haley or Minuchin.

In the 1970s, the Milan Group addressed a difficulty often encountered when the family was referred by another therapist. When these families failed to improve, the team surmised that the original treatment had reached an impasse because the referring therapist had been inducted into the system. Further, the family’s loyalty to the original therapist hindered progress in the new treatment context. As a result, the Milan team began “honoring” the original therapist by including him/her in the treatment and positively connoting the homeostasis. They might, for example, thank the therapist for helping the family by failing to produce change, leading to shock and surprise in both the therapist and the family. The team would exclude the referring therapist from their conferencing, treating him/her as a member of the family in treatment. If the team cautioned the family not to change, the referring therapist was free to disagree. The family maintained its loyalty to the original therapist, siding with him/her to prove the consulting team wrong by changing. This process, known as meta-consultation, solidified the use of the observing team. When therapy became stuck, even the Milan therapist conducting the session with the family was not exempt from being included in the positive connotation.

More recently, the model has been characterized as a second-order cybernetic systems approach. First-order cybernetics “pictured a family system in trouble as a homeostatic machine … with symptoms playing an important part of maintaining the homeostasis” (p. 14). The therapist and family were viewed as separate entities. In contrast, second-order cybernetics conceives of the therapist and family as one unit. As von Foerster noted, “[T]he observer enters into the description of that which is observed in such a way that objectivity is not at all possible … There is no such thing as a separately observed system” (von Foerster, 1974, p. 14). Accordingly, it is misleading to conceptualize the family as a separate entity; rather it is “better to think of the treatment unit as a meaning system to which the treating professional is as active a contributor as anyone else” (Boscolo et al., 1987, p. 14). The system does not create the problem; the problem creates the system.

Other changes have since evolved within the Milan approach. Newer interventions reflect a greater emphasis on neutrality. Formerly, the closing statement of a session included a statement of paradox or so-called sacrifice intervention, in which the symptomatic person was characterized as working in the service of homeostasis. Such an intervention helped overcome resistance by provoking rebellion against the symptom. While the family often improved through the use of sacrifice interventions, individual members might feel guilty or blamed. More recently, paradox is used less frequently, and the messages are more neutral, in that they place “all the behaviors related to a problem in the service of a shared premise, value, or myth” (Boscolo et al., 1987, p. 16). In this way, no one in the family feels blamed, and the message “elucidates the double-level bind” (p. 16).

Additionally, the team has modified the positive connotation technique. The positive connotation of a symptom implied that the symptom was needed by the family and, therefore, was good. However, families experienced their problems as distressing, and the problem’s characterization as good could be perceived as mocking. The technique was reframed and the team gradually shifted toward the use of logical connotation. “There is no need to say that a problem is useful, beneficent, or functional—only that people have gotten used to it and that such habits are hard to break” (Boscolo et al., 1987, p. 16). The symptom’s development is neither good nor bad but understandable, given the context.

The team has also altered its use of rituals. “A ritual is an ordering of behavior in the family either on certain days … or at certain times” (Boscolo et al., 1987, p. 4). Earlier use of rituals entailed amplifying a symptom with the goal of disrupting it (e.g., having a family solemnly thank the symptomatic member each day for having the symptom). More recently, the ritual consists of simultaneous conflicting directives. For instance, a mother is asked to be simultaneously wife to her husband and mother to her daughter. But on even days she is told to be wife to her husband and on odd days, mother to her daughter. In general, Boscolo and Cecchin believe that interventions should do no more than perturb the system so that it can react on its own terms. Interventions, then, are not geared toward any particular outcome but rather to jog the system to find its own solution, often in ways that are surprising to all.

The Milan team has been influenced by cognitive biologists (e.g., Humberto Maturana and Heinz von Foerster) and radical constructivists (Ernst von Glasersfeld). Greater validity is given to the concept that reality is a social construct rather than based on objectively “real” external events. “Ideas, beliefs, myths, values, perceptions, fantasies, and other internal productions” (p. 19) therefore assume greater importance. Families are understood to unconsciously construct maps or premises about their world to help them organize their experience. The therapist looks for the premise that attaches behaviors to a problem and tries to articulate it to the family. The family may then shift its premise and change behaviors accordingly. Thus, for the Milan team, meaning is primary, and the resulting behavioral change would be described, in MRI terms, as stemming from a second-order change.

In the 1980s, Boscolo and Cecchin split from Selvini and Prata, becoming the Milan Associates. While Selvini and Prata focused mainly on research, the Milan Associates focused on training, which significantly shaped the evolving treatment model. Trainers worked in small private clinics, while trainees tended to work in public settings, where the ideas espoused by family systems therapists were often met with hostility. Families treated in these settings were also resistant to the idea that the whole family needed treatment, since the existing models suggested that only the person with the symptoms needed treatment. Thus, the trainees found themselves dealing not only with families, but with the treatment milieu as well. This experience highlighted the importance of the larger context, or the “significant system” in which the treatment occurs. The schools, courts, clinics, and cultures that the therapists and families are involved with may all have an effect on the treatment. In response to the feedback from the trainees and the systems, the Milan Associates decided to “think of themselves not as family therapists but as systems consultants” (p. 24).

♦♦♦♦♦

Selvini Palazzoli, M., Boscolo, L., Cecchin, G., & Prata, G. (1980). Hypothesizing—circularity—neutrality: Three guidelines for the conductor of the session. Family Process, 19(1), 3–12.

This article outlines the three principles—hypothesizing, circularity, and neutrality—developed by the Milan Group for conducting therapy. The information that is known about a family prior to the first meeting forms the basis for an initial hypothesis about the family process: names, ages, jobs, education, referral source, reported problem, and the like. According to the authors, the hypothesis is an

unproved supposition tentatively accepted to provide a basis for further investigation (p. 5) based upon the information [the therapist] possesses regarding the family … The hypothesis establishes a starting point for his [or her] investigation as well as his [or her] verification for the validity of this hypothesis based upon specific methods and skills. If the hypothesis is proven false, the therapist must form a second hypothesis based upon the information gathered during the testing of the first (p. 4).

The hypothesis, then, helps the therapist decide what he/she might particularly attend to in the first session and leads to uncovering essential information. Subsequently, the data gathered in the first session may rule out the initial hypothesis completely and form the basis for an alternative hypothesis. Making hypotheses requires that therapists take account of all observations and organize them into a meaningful construct. Thus, the action of hypothesis building is experimental. New information leads to confirmation or rejection of the working hypothesis followed by the formulation of refined or altogether new hypotheses.

The value of a hypothesis is not tied to whether it is true or false, but whether it is more useful or less useful as a guide to furnishing the therapist and family with new information. The hypothesis functions as a discipline for the treatment and a guide to gathering new information. It helps the therapist track the interactional patterns.

Working through a hypothesis keeps the therapy from falling into disorder and muddle. Systematic, active hypothesis testing helps counter entropy. Entropy in a system refers to “the disorder, disorganization, lack of patterning, or randomness, [and a] decrease in entropy can be taken as a measure of the amount of information” (p. 6).

Circularity, according to Selvini Palazzoli, Boscolo, Cecchin, and Prata, refers to “the capacity of the therapist to conduct his [or her] investigation on the basis of feedback from the family in response to the information he [or she] solicits about relationships and, therefore, about difference and change” (p. 8). The Milan team defines circularity as the ability to obtain authentic information from the family. Bateson’s construct—“that all knowledge of external events is derived from the relationship between them” (p. 8)—suggests that in order for the therapist to obtain authentic information, every member of the system must describe his or her view of the relationship between other dyads of the system. For example, a wife may be asked how she sees the relationship between her husband and son.

Furthermore, resistance is lessened if one part of the system comments on another. In this way, circular questions unearth a wealth of information about triadic relationships in the family and effectively break the rule(s) about secrets in dysfunctional families.

The Milan team offers other suggestions for gathering information:

The more information gained from each subsystem member, the larger the field of observation. Once a relatively clear picture is assembled, the field is enlarged to include relationships with extended families and families of origin.

The Milan therapists maintain a neutral relationship with each family member. By the end of the session, family members might have a sense of what the therapist is like but should have no sense that he/she has sided with anyone or made any judgments about the entire family. Circular questions shift the from one member to the next as each is asked to comment on the other relationships within the family. The more the therapist emphasizes curiosity and information gathering, the less likely he/she is to make moral judgments. Therapists make a conscious effort to prevent family members from forming coalitions with him/her. The therapist is effective only insofar as he/she is able to remain at a different level—a metalevel—relative to the family.

¹ According to Nichols & Schwartz (1998), as strategic therapy models have evolved, the use of paradoxical interventions has declined in view of the necessary use of deception.