Milan systemic and post-Milan therapy

1. Introduction

Version followed: three named strands, kept apart throughout. (A) Classic Milan = the four-person team of Mara Selvini Palazzoli, Luigi Boscolo, Gianfranco Cecchin and Giuliana Prata (1971 to 1980), as set out in Paradox and Counterparadox (Italian 1975, English 1978) and the 1980 guidelines paper. (B) Post-Milan (Boscolo & Cecchin) = the Centro Milanese di Terapia della Famiglia (CMTF) line after the 1980 split: Boscolo, Cecchin, Hoffman & Penn (1987), Cecchin (1987) on curiosity, Cecchin, Lane & Ray (1992, 1994) on irreverence and prejudices, Boscolo & Bertrando (1993, 1996) on time and individual systemic therapy, plus Karl Tomm's interventive interviewing (1987 to 1988), the most widely taught operationalisation. (C) Selvini/Prata research line = the invariant prescription (1980s) and the psychotic family games model (Selvini Palazzoli, Cirillo, Selvini & Sorrentino, 1989). Where a technique belongs to only one strand it is labelled (A), (B) or (C). [source: Springer encyclopedia entry on Cecchin; Barbetta 2017, CMTF Connessioni; Wikipedia, Selvini Palazzoli; Paidós synopsis of Los juegos psicóticos en la familia]

What it is. A school of family therapy, not a single protocol. A psychoanalytically trained group in Milan turned to Bateson's cybernetics and the Palo Alto (MRI) communication work and began treating families "in schizophrenic transaction" and families of young people with self-starvation as systems with their own rules or "games". The team worked collectively behind a one-way mirror; the therapist in the room spoke for the team. [source: Wikipedia, Selvini Palazzoli; Sommella 2017 review, Terapia Familiare; Aguirre Peña et al., Universidad Católica Luis Amigó review]

How change happens. Classic Milan (A): symptoms are part of a family game that maintains the system's balance. The team positively connotes that game, then uses counter-paradox (prescribing no change, prescribing the symptom) or rituals to make the game impossible to continue unchanged. Post-Milan (B): change comes when beliefs and meanings change, because circular questions bring forward new information (differences, connections between people). The interview itself is now the main intervention, and the end-of-session message matters less. [source: AACAP Systemic Family Therapy training slides 2022; Carr 2000 on Tomm; Sommella 2017] Strand (C) returns to a directive, researcher stance. A standard series of prescriptions is given to the parents to break the family game and to map how it works. [source: Paidós synopsis]

No licensing. No certification is required to use the methods. Training is offered by systemic training institutes, notably CMTF (Milan) and the Scuola Mara Selvini Palazzoli (Milan). [source: Springer entry; Wikipedia] [uncertain: present-day course structures]

2. Framework

Format (A, classic).

Whole family invited, usually nuclear family plus any relatives living in the household. Often a male-female pair of co-therapists in the room, with the rest of the team behind a one-way mirror. [source: Aguirre Peña et al.; Barbetta 2017] [uncertain: the co-therapy-pair detail is from Paradox and Counterparadox]
First contact by telephone. One team member takes the call and records the referral, who lives at home, who referred, and how the problem is described. This record is used for the first hypothesis. [source: Aguirre Peña et al., citing Selvini et al. 1975]
Five-part session, described as highly ritualised [source: Aguirre Peña et al., citing Bertrando & Toffanetti 2000]. The five parts are: (1) presession, where the team builds hypotheses from the referral data and previous sessions; (2) session/interview, about an hour, in which the therapist tests hypotheses by circular questioning while the team may call the therapist out or phone in; (3) intersession, where therapist(s) and team confer away from the family and form a systemic hypothesis; (4) intervention, where the therapist returns and delivers a short message or prescription, then the family leaves without discussion; (5) postsession, where the team discusses the family's reaction and writes notes. [source: Aguirre Peña et al. for presession, intersession/final intervention, prescription; part names follow standard Milan usage] [uncertain: exact naming of parts 2 and 4 varies by text]
Spacing and length. Sessions were about a month apart and a course was about ten sessions, a format called "long brief therapy". The interval was a deliberate choice: it gave prescriptions time to work through the system. [uncertain: from Paradox and Counterparadox and Boscolo et al. 1987; neither is drawn on, and no source gives figures]

Format (B, post-Milan). Same team/mirror structure in training and clinic settings, but more flexible. Teams may join the room, use reflecting-team conversations (Andersen) where the family hears the team, see individuals and couples, and vary the interval. Often there is no end-of-session prescription, only a summary or reflection, or nothing. [source: Barbetta 2017; Springer entry ref. Cecchin 1993 "From strategizing to nonintervention"] [uncertain: interval norms in B]

Format (C, invariant prescription). The whole family is seen first. Then the parents are seen alone, and the children are dismissed from therapy. The parents are given a fixed series of prescriptions across sessions and report back on each. The research cohort was 149 families who received the same series over eight years. [source: Paidós synopsis for the 149 families, eight years and the fixed series] [uncertain: the session-by-session sequence is from Prata 1990 and Selvini Palazzoli 1986/1989]

Phases and transition criteria. The sources describe no formal staged protocol. In practice, working phases are as follows. [synthesis]

Referral and hypothesis. Telephone intake, decide who to invite, deal with the referring person. Move on once a working hypothesis covers every household member and the referrer.
Exploratory interviewing. Circular questioning across sessions 1 to 3. Move on once the team can describe the relational pattern around the symptom in a way that includes everyone and blames no one.
Intervention. Positive connotation, plus a prescription or ritual (A), or continued reflexive questioning (B), or the prescription series (C). Judge each step by the family's response, reported next session; an intervention is kept or dropped depending on what it produces.
Ending. Ending is often proposed by the team, sometimes paradoxically, for example a cautious ending while warning against too fast a change. End when the symptom has gone or become non-central and the family is describing itself differently. [uncertain: no source states formal ending criteria]

3. Guiding principles

Hypothesising. A hypothesis is a working supposition, neither true nor false, only more or less useful, and it must include every family member and their relationships. In practice: arrive with a relational hypothesis, test it, and drop it if the answers do not fit. [source: AACAP 2022; Aguirre Peña et al.; Selvini Palazzoli et al. 1980, via secondary sources]
Circularity. The therapist steers by the feedback the family gives to the therapist's questions about relationships, differences and change. In practice: let each answer choose the next question, and ask about relationships, not traits. [source: AACAP 2022; Aguirre Peña et al. citing Selvini 1990, Tomm 1988]
Neutrality. Neutrality refers to the pragmatic effect of the therapist's whole behaviour, not to an inner attitude. The family should experience the therapist as allied to everyone and to no one. In practice: give balanced time, ask everyone the same question in turn, and do not side with any member or any explanation. [source: AACAP 2022 citing Fleuridas et al. 1986; Aguirre Peña et al. quoting Selvini et al. 1980]
Curiosity (B). Cecchin restated neutrality as curiosity: a stance of interest in many possible stories that keeps the therapist from falling in love with one hypothesis. In practice: when you feel sure, ask yourself what else might explain it. [source: AnnaFreud/AMBIT abstract of Cecchin 1987]
Positive connotation. Every member's behaviour, the symptom included, is described as serving the family's cohesion or protection, with no one blamed. In practice: give credit for the logic of the pattern before suggesting any change. [source: Sommella 2017; Aguirre Peña et al. citing Boscolo et al.]
The family as a self-regulating system with rules or games. In practice: treat the symptom as one move in a pattern, and ask what it does for the relationships. [source: Sommella 2017; Carr 2000 (the problem as a solution to another difficulty, citing Campbell, Draper & Crutchley 1991)]
The team as the unit of therapy. Interventions come from the team, and the therapist acts as its messenger. In practice: use breaks, and use the team as a source of perspective and a check on your own drift. [source: Aguirre Peña et al. citing Bertrando & Toffanetti 2000]
Watch the wider system, including the referrer. The referring person or other professionals may be part of the family's homeostasis. In practice: map the referral pathway before the first session. [source: DiNicola 1984 abstract, Minkowska site; Selvini Palazzoli et al. 1980 "The problem of the referring person"]
Everything the therapist does is an intervention (Tomm). In practice: choose questions on purpose, knowing every question has effects. [source: Carr 2000]
Strategizing (Tomm's fourth guideline). The therapist decides deliberately how to proceed, cycling between circularity, hypothesising, strategizing and neutrality. [source: Obsidian study notes on Tomm 1987 Part I; Carr 2000]
Irreverence (B). Hold no theory, technique or instruction, including your own model, too seriously, and be respectfully ironic. In practice: if a method is not working, step outside it. [source: Barbetta 2017; Springer entry refs]
Prejudices are unavoidable and should be made visible (B). In practice: name your own assumptions to the team, and sometimes to the family. [source: Springer entry ref. Cecchin, Lane & Ray 1994; Aguirre Peña et al. (hypotheses include the therapist's biases)] [uncertain: book]
Time and meaning (B). Present problems are tied to stories about the past and expectations of the future. In practice: ask about past, present and future, and in the hypothetical mood. [source: Barbetta 2017 on Boscolo & Bertrando]
Transparency (A). The review quotes Selvini on keeping nothing hidden from the patient. Note the tension with strategic paradox; in modern practice, prefer honest positive connotation to deceptive manoeuvres. [source: Sommella 2017] [synthesis]

4. Intervention inventory

Sources for this section: AACAP 2022; Aguirre Peña et al.; Carr 2000; Sommella 2017; Barbetta 2017; the Obsidian Tomm notes; the Paidós synopsis; DiNicola 1984 abstract. Specific procedural detail drawn from primary texts that are not drawn on is tagged [uncertain]. All example wording is illustrative and is not quoted from the founders.

4.1 Telephone intake and first-contact record (A, B)

Purpose / when: Before session 1, to gather data for the first hypothesis and decide who to invite.
Steps: (1) A team member takes the call and records who is calling, who lives in the household (names, ages, occupations), the problem in the caller's own words, when it began, who referred and why now, and previous treatments. (2) Note any contact the referrer has with the family. (3) Invite everyone living in the household to session 1. (4) Bring the record to the presession. [source: Aguirre Peña et al., citing Selvini et al. 1975] [uncertain: fields of the original chart]
Example language (illustrative): "For the first meeting we usually like to see everyone who lives at home, because everyone sees the situation from a different angle."
Watch for: Who calls and who is "too busy" to attend; pressure to see only the identified patient. These are data for the hypothesis.

4.2 Presession hypothesising (A, B)

Purpose / when: Before every session, to arrive with a direction rather than a blank, and to avoid being organised by the family's own script of who is mad or bad.
Steps: (1) The team reads the record and notes from the last session. (2) Each member offers a hypothesis that links the symptom to relationships among all members, including absent ones and the referrer. (3) Choose one or two to test. (4) Plan the opening questions that will test them. [source: AACAP 2022; Aguirre Peña et al.]
Example (illustrative hypothesis): "Since the elder son left home, the daughter's refusal to eat keeps the parents talking to each other through her."
Watch for: Linear or blaming hypotheses ("the mother is overinvolved") are not yet systemic. Keep revising each one until every member has a place in it.

4.3 Circular questioning (A, B; the core technique)

Purpose / when: Throughout every interview, to test hypotheses, bring out differences and connections, and give the family new information about itself. [source: AACAP 2022; Aguirre Peña et al.]
Steps: (1) Start from the problem as presented. (2) Ask about relationships and behaviour, not inner states or traits. (3) Use the family's answers to choose the next question (circularity). (4) Ask the same question of each member in turn. (5) Rotate through the question types below. (6) Use past, present and future tenses, and hypothetical questions. [source: AACAP 2022 citing Fleuridas et al. 1986; Penn 1982]
Question types (Fleuridas, Nelson & Rosenthal 1986 classification as given in AACAP 2022, plus Penn 1982; examples illustrative):
- Problem-definition and difference questions: "When did the worry about Anna start, and what was different between your parents back then?"
- Sequence of interaction: "When Tom shouts, what does Dad do? And then what does Mum do?"
- Comparison and ranking (classification): "Who is most upset when she won't eat? Who next? Who least?"
- Triadic questions, asking one person about the relationship of two others in their presence (sometimes called "gossiping in the presence"): "Sara, how do you think your mother and your brother get along when Dad is away?"
- Agreement/disagreement: "Does your father agree with your mother's view of this?"
- Explanation/meaning: "Why do you think Grandma gets involved at those moments?"
- Before/after and hypothetical/future: "If she started eating again tomorrow, who would notice first? What would change between your parents?" [source: AACAP 2022; Carr 2000 examples "Who else worries?"]
Watch for: It is working when members show surprise, connect things in new ways (for example, a husband sees that his depression brings on his wife's complaints, not only the reverse) and stop rehearsing blame [source: AACAP 2022]. It is not working when the session becomes an interrogation. Balance questions with statements and show your human presence [source: AACAP 2022].

4.4 Maintaining neutrality in session (A, B)

Purpose / when: Every session, to avoid joining a coalition and so losing the ability to influence the whole system.
Steps: (1) Give each member roughly equal time. (2) Ask everyone the same key question. (3) Do not accept any member's explanation as the true one. (4) After the session, the team checks whether the family experienced the therapist as siding with someone. (5) If so, correct this next time by questioning the favoured member equally. [source: AACAP 2022; Aguirre Peña et al.]
Example language (illustrative): "That's one view. I'd like to hear how your sister sees it."
Watch for: At the end of a session, family members should be unable to say whose side the therapist was on. Neutrality toward people is never neutrality toward violence or abuse (see 5). [synthesis]

4.5 Team behind the mirror and mid-session consultation (A, B)

Purpose / when: To keep a meta-position the in-room therapist cannot hold alone, and to catch therapist drift.
Steps: (1) Obtain the family's informed consent to observation and recording, and introduce the team. (2) Agree how the team signals: a knock, a phone, or calling the therapist out. (3) Team calls are kept short: a new hypothesis, a question to ask, or a warning that the therapist is being drawn in. (4) The therapist returns and folds the suggestion into the interview. [source: Aguirre Peña et al.; Barbetta 2017] [uncertain: signalling conventions]
Without a team (adaptation): Use a short self-consultation break, a single colleague, or recorded review. [synthesis]
Watch for: A family asking about or addressing the team is useful material. Note who is curious and who is anxious.

4.6 Intersession team discussion (A, B)

Purpose / when: Near the end of the session, to reach a systemic hypothesis about what the symptom means and to design the closing intervention. [source: Aguirre Peña et al.]
Steps: (1) The therapist leaves the family about 10 to 20 minutes before the end. (2) The team pools observations and refines the hypothesis. (3) It chooses an intervention type: positive connotation, prescription, ritual, letter, or (in B) a reflective summary or none. (4) It drafts the exact wording. (5) It agrees how the therapist will deliver it and then end without debate. [uncertain: timings]
Watch for: Team members competing for "the" hypothesis. Irreverence and curiosity apply to the team too (see 4.17).

4.7 The end-of-session intervention or message (A; optional in B)

Purpose / when: To introduce a new frame or task at a point where the family cannot immediately argue it away.
Steps: (1) Return and say the team has talked. (2) Read or deliver a short statement that positively connotes the pattern (4.8), optionally followed by a prescription (4.9 to 4.11). (3) Do not invite discussion. Close the session and book the next one. (4) In the postsession, note each member's immediate reaction. [source: Aguirre Peña et al.; Carr 2000] [uncertain: the "no discussion" convention is from Paradox and Counterparadox]
Example language (illustrative): "The team was impressed by how much each of you is doing to keep the family together at a difficult time. We think it would be unwise to change anything quickly. We'll see you in four weeks."
Watch for: Reports at the next session. Families often mention the message unprompted. Disqualifying it ("what you said was nonsense") is still a response to work with.

4.8 Positive connotation (A, core; B in softened form)

Purpose / when: Before any prescription, and whenever blame dominates. It puts all members on the same level, removes the bad/mad split, and frames the symptom as serving the family. [source: Sommella 2017; Aguirre Peña et al.]
Steps: (1) Identify each member's part in the pattern around the symptom. (2) Connote each part positively by its function for the family, for example protecting someone, holding the family together, or keeping a loss from being felt. Connote everyone's part, not only the patient's. (3) Link the parts into one statement. (4) Do not suggest that anyone should stop. [source: Sommella 2017; Carr 2000]
Example language (illustrative): "By staying ill, Marco gives his parents a reason to stay close. His mother protects him, and his father keeps his distance so as not to compete. Each of you is doing this out of care for the family."
Watch for: It is working when blame drops and the family shows puzzlement or relief. Never positively connote abuse or violence itself. Connote the concern or care behind the attempts at solution instead. [synthesis]

4.9 Paradoxical prescription: prescribing the symptom or no change (A)

Purpose / when: In classic Milan, with rigid systems that disqualify direct advice. The symptom is prescribed, so it can no longer be spontaneous. [source: Sommella 2017] (Note: Aguirre Peña et al., citing Boscolo et al., say the Milan team did not prescribe the symptom in the later model [uncertain: the sources reflect different periods].)
Steps: (1) Deliver a positive connotation. (2) Prescribe continuing the pattern, or warn against change for now, for the good of the family. (3) Frame the prescription as a team decision. (4) At the next session, explore what happened without praising or blaming. [source: Sommella 2017] [uncertain: further procedure is in Paradox and Counterparadox]
Example language (illustrative): "Given how much this protects everyone, we suggest that for now nobody try to change how things are at mealtimes."
Watch for: Do not use this where symptom continuation carries risk: self-starvation at medical risk, self-harm, violence, or substance use. In those cases, use positive connotation of concern plus safety planning (see 5). [synthesis]

4.10 Family rituals (A)

Purpose / when: To give the family a prescribed sequence of acts, at set times and places and done by all, which creates a corrective experience that words cannot. [source: Sommella 2017] [uncertain: details in Selvini Palazzoli et al. 1977 "Family rituals"]
Steps: (1) Design a ritual from the hypothesis that enacts a new rule, for example a family meeting where each speaks for a set time without interruption, or a ritual marking a loss the family has not mourned. (2) Specify who, when, where, how long, and what each person does. (3) Ask for it to be done exactly as described. (4) Ask for an account of how it went at the next session. [synthesis on design; Sommella 2017 on the format]
Example language (illustrative): "On each of the next four Thursday evenings, from 8 to 8:30, the family will meet at the kitchen table. Each person will have five minutes to say anything about the week while the others listen without answering."
Watch for: Whether it was done, partly done or forgotten, and by whom. Each is information about the system.

4.11 "Odd days and even days" ritual (A)

Purpose / when: Classic use is where parents disqualify each other's parenting and the child moves between them. The ritual separates the parents' competing actions in time so each acts without interference and the triangulation becomes visible. [source: DiNicola 1984 abstract lists Selvini et al. 1978, "Odd days and even days"] [uncertain: the following steps are from the 1978 paper]
Steps: (1) Agree days, for example one parent on Monday, Wednesday and Friday and the other on Tuesday, Thursday and Saturday. (2) On their days, that parent alone decides how to respond to the child's problem behaviour, and the other parent behaves as if absent from that decision. (3) On the remaining day, everyone behaves spontaneously. (4) The off-duty parent writes down any departure from the ritual, without commenting at the time. (5) Review the notes at the next session.
Example language (illustrative): "From Monday, on odd days your mother alone will decide what happens when Luca refuses school, and your father will act as though he isn't there. On even days it's the other way round."
Watch for: Check the safety of each parent's approach before prescribing. Do not use it where one parent is violent or neglectful, or where separating authority would endanger a child. [synthesis]

4.12 Working with the referring person (A, C)

Purpose / when: When the referrer is a professional, relative or family friend who has taken on a homeostatic role in the family, such as rescuer, go-between or substitute parent. If the referrer is left out, the therapy can stall. [source: DiNicola 1984 abstract; Selvini Palazzoli et al. 1980 "The problem of the referring person"]
Steps: (1) At intake, map the referrer's relationship with the family: how often they are in contact, what they do, and how they would be affected if the problem resolved. (2) If the referrer is closely involved, include them in the hypothesis. (3) Consider inviting them to an early session, or positively connoting their role in the closing message. (4) Avoid the therapist becoming one more helper in the referrer's chain. [uncertain: the specific moves come from the 1980 paper; steps 1 and 4 are synthesis]
Example language (illustrative): "Dr Rossi has clearly been a great support. We wonder what it would mean for her and for you if Gianni no longer needed so much help."
Watch for: Sudden drop-out after the referrer withdraws or objects. Repeated re-referral to new services.

4.13 Session spacing as an intervention (A)

Purpose / when: A long interval (about monthly) gives the system time to reorganise around a prescription and lessens dependence on the therapist. [uncertain: interval rationale from Paradox and Counterparadox and Boscolo et al. 1987]
Steps: Set the next date at the end of each session. Lengthen or keep the interval rather than shortening it in response to family pressure, unless risk requires closer contact. [synthesis on the risk exception]
Watch for: Crises between sessions. Have a clear safety route, and do not let a calendar rule override risk management.

4.14 Postsession discussion (A, B)

Purpose / when: After the family leaves, to record the family's reaction to the message, update hypotheses, and look at the team's own process.
Steps: (1) Each member reports what they saw in the family's reaction to the message. (2) Revise the hypotheses. (3) Note any sign that the therapist or team was drawn into the family game. (4) Write the session record. [source: Aguirre Peña et al.]

4.15 Interventive interviewing: Tomm's four question types and strategizing (B)

Purpose / when: To choose questions deliberately by their intent. The intent is either orienting (finding out) or influencing, and the assumptions are either lineal or circular. [source: AACAP 2022; Carr 2000]
The four types: Lineal questions orient the therapist with cause-and-effect questions ("What brings you here?"). Circular questions explore patterns that connect, with the therapist as explorer. Strategic questions correct by leading ("Why don't you tell him how you feel?"), with the therapist as teacher; they are powerful but can harm the alliance. Reflexive questions are facilitative and help the family use its own resources, with the therapist as coach. [source: AACAP 2022; Carr 2000]
Reflexive subtypes (Tomm 1987, Part II): future-oriented, observer-perspective, unexpected context-change, embedded-suggestion, normative-comparison, distinction-clarifying, questions that introduce a hypothesis, and process-interrupting. [uncertain: the Part II paper is not drawn on; the list is given from the published titles of the categories and should be checked]
Strategizing: Tomm adds strategizing as a fourth guideline. During the session, the therapist cycles through observing feedback (circularity), hypothesising, deciding how to proceed (strategizing) and staying open (neutrality), then back again. [source: Obsidian study notes on Tomm Part I]
Steps: (1) Open with lineal and circular questions to orient. (2) As a hypothesis firms up, move to reflexive questions. (3) Keep strategic questions for moments when you have a clear reason, and repair the alliance afterwards. [source: Carr 2000; AACAP 2022]
Example language (illustrative), reflexive: "Suppose in a year from now the two of you were arguing less. What would your son be doing differently?" (future-oriented); "If your mother were here, what would she say is keeping you stuck?" (observer-perspective).
Watch for: Novices can freeze from over-planning every question. Keep the relationship first. [source: Carr 2000]

4.16 Tracking openings and key words (B)

Purpose / when: To find entry points for hypotheses in the family's own language.
Steps: (1) Listen for repeated or emotionally loaded words, and for things said in passing ("since grandad died..."). (2) Note them. (3) Return to them with circular questions about who uses the word, when, and with what effect. [source: Aguirre Peña et al., citing Boscolo et al. on "aperturas" (openings)] [uncertain: the key-word technique is described in Boscolo & Bertrando's books]
Example language (illustrative): "You said 'betrayal' twice. Who in the family uses that word most?"

4.17 Curiosity (B)

Purpose / when: When the therapist or team becomes certain, bored, or committed to one hypothesis.
Steps: (1) Notice certainty. (2) Generate at least two other descriptions of the pattern. (3) Ask questions that could disconfirm the favoured hypothesis. (4) Value the family's own explanations as further stories rather than errors. [source: AMBIT abstract of Cecchin 1987; Aguirre Peña et al.] [synthesis on steps 2 and 3]
Watch for: Curiosity turns into intrusiveness when it ignores distress. Pace it to the family.

4.18 Irreverence (B)

Purpose / when: When the therapist is trapped by a model, an institutional mandate, or a rule from the family. Respectful irony and openness to switching method keep the therapist free to act. [source: Barbetta 2017; Springer entry refs. Cecchin, Lane & Ray 1992; Cecchin 1993]
Steps: (1) Name to yourself or the team the rule you are obeying. (2) Consider acting against it in a small, safe way, for example giving advice in a non-directive model, being humorous, or changing the format. (3) Watch the effect. [synthesis]
Example language (illustrative): "I'm supposed to be asking questions, but I'm going to break my own rule and tell you what I think."

4.19 Making prejudices explicit (B)

Purpose / when: When the therapist's assumptions (cultural, theoretical, moral) shape the hypotheses. [source: Springer entry ref. Cecchin, Lane & Ray 1994; Aguirre Peña et al.] [uncertain]
Steps: (1) During pre- and postsession, list the assumptions behind each hypothesis. (2) Ask what a colleague from another background would see. (3) Where useful, share an assumption tentatively with the family and invite correction. [synthesis]

4.20 Questions about time (B; Boscolo & Bertrando)

Purpose / when: When a family is fixed on a past event or blocked in an imagined future, and when present problems seem tied to the time of the family life cycle.
Steps: (1) Ask how the problem connects to past events and stories. (2) Ask how each member imagines the future with and without the problem. (3) Use hypothetical or subjunctive questions ("if... what would...") to open possible worlds. (4) Link these back to the present. [source: Barbetta 2017 on Boscolo & Bertrando, I tempi del tempo]
Example language (illustrative): "If you imagine the family in five years, with this problem solved, who would have changed most?"

4.21 Individual systemic therapy (B; Boscolo & Bertrando 1996)

Purpose / when: When only one person attends, or when the family is not available.
Steps: Apply hypothesising, circular and reflexive questions, and time questions to one client, asking about significant others as if they were present ("What would your father say...?"). Treat the client's inner voices and relationships as a system. [uncertain: the book Terapia sistemica individuale; procedure inferred from strand B principles]

4.22 Reflecting-team adaptation (post-Milan derivative)

Purpose / when: To open the team's discussion to the family.
Steps: (1) Partway through, swap rooms or lights so the family watches and hears the team talk. (2) Team members speak tentatively to one another, not to the family, offering several views. (3) The family then comments on what they heard. [source: Barbetta 2017 (Andersen's reflecting team as breaking the mirror's separation)] [uncertain: procedure from Andersen's work; this belongs to Andersen rather than to the Milan authors]

4.23 Invariant prescription (C: Selvini Palazzoli & Prata)

Purpose / when: Developed in research with families of children diagnosed with psychosis or severe anorexia. The same series of prescriptions is given to every family, to strengthen the parental couple's boundary and to make the family game visible by the reactions it provokes. [source: Paidós synopsis (149 families, eight years, fixed series)] [uncertain: all steps below are from Selvini Palazzoli 1986 and Prata 1990]
Steps (as reported in secondary literature): (1) After whole-family sessions, dismiss the children with thanks and continue with the parents alone. (2) Prescribe that the parents keep the content of their sessions secret from children and relatives. (3) Prescribe evenings out together without warning, leaving only a short note. (4) Extend these later to longer absences, such as a weekend. (5) Each parent keeps a private notebook recording each family member's reactions. (6) Review the notebooks at the next session and use them to map the game.
Example language (illustrative): "What happens in these sessions stays between the two of you. If anyone asks, you can say the therapist told you to keep it private."
Watch for: Do not use this where a child or adolescent would be left unsafe, where there is current self-harm or medical risk, or where the couple relationship involves coercion or violence. Prescribed secrecy must never cover abuse. Get informed consent from the parents about the method's rationale. [synthesis]

4.24 Mapping the family game (C)

Purpose / when: As a hypothesis framework for families where a young person has severe symptoms. The symptom is linked to a stalemate in the parents' relationship, the child's involvement in it, and subsequent reversals of alliance. [source: Paidós synopsis (games as metaphor for recurring interaction processes); Sommella 2017 (covert alliances, parental couple unable to lead)] [uncertain: the six-stage model in Family Games (1989)]
Steps: (1) Reconstruct the history of the parental relationship before the symptom. (2) Ask how the child became involved and with whom they sided. (3) Identify the turning point when the expected ally did not back them. (4) Explore how the symptom now works in the game. (5) Use this map to guide questions and, in later practice, to combine individual and family sessions. [synthesis on use; Sommella 2017 on the later combined individual-and-family work, Ragazze anoressiche e bulimiche]

4.25 Consultation to larger systems (A, B)

Purpose / when: In organisations, schools, hospital wards and inter-agency networks where the problem is held between professionals. [source: Wikipedia (The Hidden Games of Organizations, 1986)]
Steps: Apply hypothesising, circular questions to staff about one another, and a positively connoted closing message to the professional network as the client system. [uncertain: specific procedures]

5. Handling problems

Signals from the session itself (Tomm). If the atmosphere feels constrained or oppositional, move back toward neutrality. If the session is dull, increase hypothesising and strategizing. If sessions are unproductive despite clear hypotheses, return to circularity and ask fresh questions about difference. Prolonged engagement can shift your stance, for example caring circularity sliding into naivety with a deceptive family, so monitor your own feelings in session. [source: Obsidian study notes on Tomm Part I]

Stuck family, the same story every session. Return to presession and generate new hypotheses that include absent members, the referrer and the therapist's own place in the system. Use curiosity (4.17) and time questions (4.20). If the team is attached to one idea, use irreverence (4.18). [source: Cecchin 1987 abstract; Barbetta 2017] [synthesis on order]
"Resistance": the family rejects the message or does not do the task. In the Milan frame, non-performance and disqualification are information about the system, not defiance to overcome. (1) Do not scold or repeat the instruction. (2) Ask circular questions about who wanted to do it, who didn't, and what happened when it was not done. (3) Revise the hypothesis. (4) Consider positively connoting the decision not to change. [uncertain: the classic stance appears in Paradox and Counterparadox] [synthesis on steps]
Members absent or refusing to attend. Ask circular questions about the absent person ("What would your father say about this if he were here?"). Consider including the absent member in the closing message or a letter. [uncertain: letters to absent members are described in classic texts] [synthesis]
Therapist drawn into a coalition. Signs: one member keeps appealing to you, you find yourself explaining or defending, or the team phones in. Respond by asking the favoured member the same questions as the others, and invite the most distant member to comment. Discuss it in the postsession. [source: AACAP 2022 (neutrality behaviours)] [synthesis]
Overwhelm or high emotion in the room. Slow the questioning, and move from triadic questions to questions addressed directly to the distressed person. Acknowledge their distress before continuing. Take a break to consult. Too many questions can hide the therapist and make the session feel inhuman, so make human statements as well. [source: AACAP 2022] [synthesis]
Rupture after a strategic question or paradox. Name your part ("I think my question came across as taking sides"), restate the positive connotation of everyone's intentions, and return to circular and reflexive questions. Strategic questions can jeopardise the alliance. [source: Carr 2000] [synthesis on repair wording]
Novice paralysis from over-planning questions. Plan two or three opening questions only, then follow the feedback. Use live supervision. [source: Carr 2000] [synthesis]
Family asks for advice or a diagnosis directly. Strand A would usually turn the question back or give a team message. Strand B allows a direct, transparent answer (irreverence), followed by a question about its effect. When information is needed for safety or medical reasons, give it plainly. [synthesis]
Problems the sources do not address (all [synthesis]):
- Risk (self-harm, suicide, medical risk in eating disorders, violence, child protection): Neutrality toward people does not extend to neutrality about safety. Conduct a standard risk assessment and safety plan outside the strategic frame, and include medical monitoring in eating disorders. Never prescribe symptom continuation or no change where the symptom is dangerous. Never use prescribed secrecy (4.23) or separation of parental authority (4.11) where it could conceal abuse or leave a child unprotected. Follow mandatory reporting duties.
- Intimate partner violence: Do not positively connote violence or prescribe joint rituals that increase contact under coercion. Screen partners separately, and consider separate work.
- Consent and transparency for paradox and prescriptions: Explain the general rationale in advance ("we sometimes suggest tasks that may seem surprising"). Prefer honest reframes that the team actually believes. Obtain explicit consent to the team, the mirror and recording.
- No team available: Use a single co-worker, a consultation break, recorded review, or later peer supervision. Keep the five-part rhythm in compressed form: brief presession notes, interview, a five-minute break, closing reflection, postsession notes.
- Cultural fit: Check whether a positive connotation or ritual fits the family's cultural and religious meanings. Make your own prejudices explicit (4.19), and ask the family to correct the frame.
- Monthly spacing in services with short contracts: Agree the number and spacing of sessions with the service, and explain the rationale for spacing to the family.

Sources

Sources:

Further reading (not drawn on) [uncertain where relied on]:

Selvini Palazzoli M., Boscolo L., Cecchin G., Prata G. (1978). Paradox and Counterparadox. Jason Aronson (Italian: Paradosso e controparadosso, 1975).
Selvini Palazzoli M. et al. (1977). Family rituals: a powerful tool in family therapy. Family Process 16.
Selvini Palazzoli M. et al. (1978). A ritualized prescription in family therapy: odd days and even days. Journal of Marriage and Family Counseling 4.
Selvini Palazzoli M. et al. (1980). Hypothesizing, circularity, neutrality: three guidelines for the conductor of the session. Family Process 19, 3 to 12.
Selvini Palazzoli M. et al. (1980). The problem of the referring person. Journal of Marital and Family Therapy 6.
Penn P. (1982). Circular questioning. Family Process 21, 267 to 280.
Tomm K. (1984). One perspective on the Milan systemic approach, Parts I and II. Journal of Marital and Family Therapy 10.
Fleuridas C., Nelson T.S., Rosenthal D.M. (1986). The evolution of circular questions. Journal of Marital and Family Therapy 12, 113 to 127.
Selvini Palazzoli M. et al. (1986). The Hidden Games of Organizations. Pantheon.
Boscolo L., Cecchin G., Hoffman L., Penn P. (1987). Milan Systemic Family Therapy: Conversations in Theory and Practice. Basic Books.
Cecchin G. (1987). Hypothesizing, circularity and neutrality revisited: an invitation to curiosity. Family Process 26, 405 to 413 (abstract only).
Tomm K. (1987, 1988). Interventive interviewing, Parts I to III. Family Process 26, 27.
Selvini Palazzoli M., Cirillo S., Selvini M., Sorrentino A.M. (1989). Family Games: General Models of Psychotic Processes in the Family. Norton.
Prata G. (1990). A Systemic Harpoon into Family Games. Brunner/Mazel.
Cecchin G., Lane G., Ray W. (1992). Irreverence: A Strategy for Therapists' Survival. Karnac.
Cecchin G., Lane G., Ray W. (1994). The Cybernetics of Prejudices in the Practice of Psychotherapy. Karnac.
Boscolo L., Bertrando P. (1993). I tempi del tempo (English: The Times of Time). Bollati Boringhieri.
Boscolo L., Bertrando P. (1996). Terapia sistemica individuale (English: Systemic Therapy with Individuals). Raffaello Cortina.
Selvini Palazzoli M., Cirillo S., Selvini M., Sorrentino A.M. (1998). Ragazze anoressiche e bulimiche. Raffaello Cortina.
Bertrando P., Toffanetti D. (2000). Storia della terapia familiare. Raffaello Cortina (cited via Aguirre Peña et al.).

← Previous · V-014Structural family therapy and person-of-the-therapist training Next · V-016 →Heidelberg systemic therapy