Clean language and symbolic modelling

1. Introduction

Version followed: James Lawley and Penny Tompkins' Symbolic Modelling, i.e. the five-stage process of Metaphors in Mind (2000) as the authors summarise it, plus their single-session "Symbolic Modelling Lite" (2011 chapter). The question set follows their "9+3" basic questions (2004) and the specialised-question list (Lawley 2023). Clean Space follows Lawley and Manea's (2017) account of David Grove's late process. Where these versions differ, the difference is named in the text.

What it is. Clean Language is a questioning method developed by the New Zealand psychotherapist David Grove (1950–2008). He developed it from the early 1980s while working with trauma clients. He noticed that their spontaneous metaphors for symptoms faded under ordinary questioning but developed when he repeated the client's exact words inside very simple questions [source: Obituary; Clean Language Revisited]. Lawley and Tompkins modelled Grove's work and built Symbolic Modelling from it. They describe it as three parts: metaphor is the medium, modelling is the method, and Clean Language is the means [source: Clean Language Revisited; L&T 2011]. It is a method and a stance, not a manualised therapy with fixed sessions. It can be used on its own or inside another psychotherapy. Grove's later methods, Clean Space and Emergent Knowledge, are related but separate [source: Obituary; Lawley & Manea 2017].

How change happens (authors' model). The client's metaphors are treated as a working model of how their experience is organised: a "metaphor landscape" of symbols, locations and relationships. Clean questions use only the client's words, plus universal notions of time, space, form and perceiver. This lets the client "self-model", meaning they discover how their own pattern works without the therapist's assumptions being added. Once the client has an outcome they want and the system has become aware of itself, change tends to emerge from within the landscape, through self-organisation. The therapist does not engineer it [source: L&T 2011; 2017 psychotherapy interview; 1999 interview]. The authors note that the therapist may never learn the literal content behind the metaphors [source: 2017 psychotherapy interview].

2. Framework

Two published structures.

A. Five-stage process (Metaphors in Mind, 2000). Stage names are from the book's contents page [source: MiM preview]. Stage aims and tests are from the authors' 1999 interview, which splits Stage 4 into two steps (identify the conditions, then encourage them) [source: 1999 interview]. Detailed stage content in the book is not drawn on beyond the preview [uncertain].

The stages are systemic, not linear. Facilitators loop back to earlier stages often [source: 1999 interview].

B. Symbolic Modelling Lite, a single-session format (2011) [source: L&T 2011]:

1. Set up: the client chooses where they and the facilitator sit.
2. Identify a desired outcome: ask the outcome question and sort the answer with PRO (Problem, Remedy, Outcome).
3. Develop the desired-outcome landscape: keep everything relevant in one perception.
4. Explore effects: ask what happens when the outcome happens, and what then happens to the earlier problems. If a new problem appears, go back to PRO.
5. Mature changes as they occur. Grove sometimes spent up to a third of a session maturing.
6. Set down: end when changes have come to rest, the landscape is coherent, no new problems appear, and the client says they need nothing more.

Session count and format. The sources give no fixed number of sessions. The phases "can take place in one session or over a whole number of sessions" [source: 1999 interview]. The authors report three demonstrations totalling over two hours using only basic questions [source: 2017 psychotherapy interview]. The published Clean Space case lasted about 40 minutes [source: Lawley & Manea 2017]. Sessions are individual and face to face, with seating that the client places. Drawings and floor space are used as needed. A co-facilitator can take over mid-session [source: L&T 2011; 2017 psychotherapy interview]. Treat each new session as starting fresh, because the client will have changed in between [source: 1999 interview]. A practical cadence for psychotherapy, such as weekly 50–60-minute sessions with a review every four to six sessions, is [synthesis].

Moving between phases. Use the tests in the table above. Move to maturing as soon as any change appears, even a small one [source: Rock and hard place; L&T 2011]. If maturing brings back the problem or doubts, treat that as information about the client's current landscape, not as failure. Go back to PRO or to modelling the pattern [source: L&T 2011].

3. Guiding principles

Sources: L&T 2011; 2017 psychotherapy interview; 1999 interview; Less is More; Clean Language Revisited; Rock and hard place; Competency guidance.

• Use the client's exact words. Repeat words, and where possible intonation and gestures, as given. Small substitutions carry hidden assumptions.
• Add only universal metaphors. Questions may introduce only time, space, form and perceiver, never your own content, metaphors or reframes.
• Ask the question the client's information wants asked. Each answer decides the next question. Follow the logic of the landscape, not your hypothesis.
• Self-modelling is the aim. The client discovers how their system works. You don't interpret it for them.
• No intention to change the client. The authors require only that the client intends to change. Your job is to keep the work connected to the client's outcome.
• Outcome orientation through PRO. Keep sorting statements into problem, remedy or outcome. Acknowledge problems without dwelling on them, and keep directing attention toward what the client would like to have happen.
• Not knowing; trust the system. Accept that you don't know where it will go. Let solutions emerge from the landscape.
• Set aside your own perceptual space. Only the client's landscape should occupy the room. In practice, notice your own pictures of the client's metaphors ("our dragon, not their dragon") and keep them out of your questions.
• Rapport with the landscape, not the person. Don't use NLP-style physical mirroring. Direct your eye line at the client's symbols, often slightly to the side, so the client attends to them rather than to you.
• Do as little as possible. Clients usually cope with long-standing patterns. Simple questions and waiting are often enough.
• Don't bail out. When the client is stuck, model the stuckness instead of switching technique. Stuckness in the room is information.
• Respect the metaphor's short life. Grove held that metaphors fade quickly unless they are handled cleanly, so develop a metaphor before you move on.
• Content-free is acceptable. You need not know the literal story. Some clients need to tell it, so listen; don't push for it.
• The client controls the process. Clients can stop at any time, and the therapist will stop whenever asked.
• Work within your competence. Coaches should not work with severe mental disturbance. Clinicians apply the method within their own scope of practice.
• Separate your stuckness from the client's. If you are lost, step back physically and go back to the basic questions.
• Mature every change. Develop any change right away and check that every part of the landscape is satisfied with it.
• Wipe the slate clean each session. Don't assume last week's landscape still holds.
• Treat landscapes as experience, not historical fact [synthesis, required for trauma work]. Never use source or time questions to search for or "confirm" hidden past events.

4. Intervention inventory

Notation: [X] means the client's exact words. Question wordings below are the published clean questions. They are short and quoted from the sources cited. Other example lines are illustrative.

4.1 Clean Language syntax and delivery (the base of every intervention)

• Purpose: Acknowledge the client's words, direct attention to part of their experience, and send them on an inner search, while adding as little of you as possible. The authors name three functions (acknowledge, orient, send on a quest) and the components syntax, vocal qualities, gestures and questions [source: L&T 2011; Less is More].
• Steps:
  1. Start with "And", then repeat a short part of the client's words, adjusting only the pronoun ("I" becomes "you").
  2. Add "and when/as" and repeat the words again. This orients attention.
  3. Ask one clean question about those words.
  4. Use the client's exact nouns and adjectives. Mirror their emphasis, sighs and gestures, including gesturing to where they placed a symbol.
  5. Voice: roughly half normal speed, slightly lower pitch, with a regular, slightly sing-song rhythm and genuine curiosity [source: Less is More].
  6. Use the full syntax when attention needs settling. Shorten it to "And [X]... what kind of [X]?" or the bare question once the client is absorbed. The authors' practice materials include drills on varying how many questions you ask per minute [source: practice activities page].
• Example language: for "I go blank", "And you go blank. And when you go blank, what kind of blank is that blank?" (pattern from Less is More, English rendering).
• Watch for: substituting synonyms, summarising, adding "you said" or "I wonder", or leaking your own metaphors (e.g. calling the client's "weight" a "burden"). It is working when the client slows down, looks into their landscape, gestures to locations and answers from inside the experience.

4.2 Desired-outcome question (opening and re-orienting)

• Purpose: Establish the client's intention, which turns modelling into change work. It is the usual opening question and the main recovery move when you are lost [source: Clean Language Revisited; L&T 2011].
• Steps: Ask it at the start. After the client describes a problem, ask "And when [problem], what would you like to have happen?" Later, re-anchor with the client's outcome in their exact words. The authors' materials include a drill on variations of this question [source: practice activities page]; the variations themselves are not described here [uncertain].
• Example language: "And what would you like to have happen?" [source: L&T 2011]; "And what would you like to have happen now?" [source: L&T 2011].
• Watch for: the answer may be a problem or a remedy rather than an outcome (see 4.3). It is working when the client states something they want to exist, in positive, perceivable terms.

4.3 PRO model (Problem, Remedy, Outcome)

• Purpose: A running classification in the therapist's mind that decides the next question [source: L&T 2011; Competency guidance].
• Steps: Sort each statement.
  • Problem (something present that the client dislikes): ask "And when [problem], what would you like to have happen?"
  • Remedy (a wish for the problem to stop, shrink or be avoided, e.g. "I want to stop suffocating"): ask "And when [remedy], then what happens?" to find what lies beyond it.
  • Outcome (a wish for something new to exist, e.g. "to be at home with myself"): develop it with "And what kind of [outcome] is that?"
  • When you echo a remedy or outcome, leave out the desire words (say "at home with yourself", not "you want to be at home with yourself") [source: L&T 2011].
• Watch for: spending long on problem detail (acknowledge, then redirect). The opposite error is jumping to an outcome while ignoring a bind; for binds, see 4.17 [source: Rock and hard place].

4.4 Attribute questions (developing form)

• Purpose: Elaborate a word or symbol until its nature is clear. Developing questions hold time still and keep attention on the current perception [source: Clean Language Revisited].
• Steps: Choose one word that carries energy or seems central. Ask "And is there anything else about [X]?" or "And what kind of [X] is that [X]?" Ask 2–4 developing questions per symbol before moving on [synthesis; the authors stress developing before moving time, Competency guidance 3c].
• Example language: "And is there anything else about that brick wall?" [source: MiM preview]; "And what kind of wound up is that?" [source: MiM preview].
• Watch for: abstract answers that go round in circles. If so, ask about location (4.5) or ask for a metaphor (4.6).

4.5 Location questions

• Purpose: Place a symbol in perceptual space. This makes it more physical and reveals relationships [source: L&T 2011; Less is More].
• Steps: "And where is [X]?"; follow with "And whereabouts?" for more precision. Note the gestured location and gesture to the same place in later questions.
• Example language: "And whereabouts on top of you?" [source: MiM preview].
• Watch for: the client pointing inside or outside the body, at a distance, or at a direction. Each of these allows specialised questions (4.10).

4.6 Metaphor question ("like what?")

• Purpose: Turn a sensory, conceptual or nonverbal expression into a metaphor. It gives a lateral shift and is the main entry into the symbolic domain [source: Less is More; Competency guidance 2b].
• Steps: "And that's [X] like what?" Use it for concepts ("confidence"), feelings ("tight") or gestures ("and that's [gesture] like what?"). To capture a whole pattern, use "And that's all like what?" [source: Rock and hard place].
• Watch for: asking it of something that is already an explicit metaphor is listed as incorrect use [source: Competency guidance]. If the client says they don't think in images, develop their literal words with attribute and location questions instead.

4.7 Relationship questions

• Purpose: Relate parts of the landscape across space [source: L&T 2011; Clean Language Revisited].
• Steps: "And is there a relationship between [X] and [Y]?" "And when [X], what happens to [Y]?" Use the second form especially after a change, to find out how the rest of the landscape responds.
• Watch for: relationships that keep the pattern in place (stage 3) and resources in the background.

4.8 Moving-time questions (sequence and source)

• Purpose: Move attention forward or back to find sequences, triggers, effects and origins [source: Clean Language Revisited; Less is More].
• Steps: Forward: "And then what happens?" / "And what happens next?" Back: "And what happens just before [X]?" Source: "And where does/could [X] come from?"
• Watch for: in trauma-related work, source questions can lead toward autobiographical memory. Follow only what the client offers. Don't press for origins or treat answers as historical evidence [synthesis].

4.9 Intention and necessary-condition questions

• Purpose: Find what the landscape itself needs for change. This is the core of stage 4 [source: Clean Language Revisited; 1999 interview].
• Steps: "And what would [X] like to have happen?" (asked of a symbol or agent); "And what needs to happen for [X]?"; "And can [X] [action]?" Ask "can" only when an agent wants or needs to do something [source: Contextually clean 2023]. When the client names a condition, develop it, then ask "And can [condition]?" If no, ask what needs to happen for that, one level at a time [synthesis built on the published questions].
• Example language: "And what needs to happen for the fish to complete the jump?" (illustrative, using the authors' case symbol).
• Watch for: a chain of conditions that reaches something the landscape can do now. The fish completing its jump in the authors' case is an example [source: L&T 2011].

4.10 Specialised (contextually clean) questions

• Purpose: Extra questions that are clean only when the client's words already presuppose them. Use the basic questions about 80% of the time [source: Less is More; Contextually clean 2023].
• Steps (condition, then question) [source: Contextually clean 2023]:
  • Plural or repeated items: "And how many [X] could there be?"
  • Human-like symbol: "And how old could [X] be?" (the authors link this to "Child Within" processes; see the trauma caution in 4.21).
  • Child symbol: "And what could [child-symbol] be wearing?"
  • Symbol with an unwanted adjective: "And what kind of [X] was that [X], before…"
  • Distance: "And how far is [location]?"
  • Container: "And is [X] (on the) inside or outside?"
  • Movement or facing: "And in which direction is/does [X]?"
  • Perceiver: "And where is [X] [seeing/hearing] [Y] from?"
  • Similar items: "And is [X] the same or different as [Y]?"
  • Separation: "And what's between [X] and [Y]?"
  • Live nonverbal shift: "And what's happening now?" / "And what just happened?"
  • Long event: "And what happens after [X]?"
  • Choice: "And what determines whether [X] or [Y]?"
  • Knowing: "And how does [Y] know [X]?"; body part active: "And is there anything [body part] knows about [X]?"
  • Thing-like experience: "And does [X] have a size or shape?" [source: Clean Language Revisited].
• Watch for: using a specialised question without its presupposition (e.g. asking "how old" of a symbol the client has not made human-like). That leads the client.

4.11 Set up (client places the room)

• Purpose: Align the client's inner and outer worlds before the work begins [source: L&T 2011].
• Steps: Before content, ask "And where would you like to be?" then "And where would you like me to be?" Let the client adjust distance and angle, and their own seat. Ask "And is that the right place?" if they hesitate [illustrative].
• Watch for: clients who defer ("wherever you like"). Allow a moment, then accept a simple choice. Notice where they place you, because it can echo landscape relationships [synthesis].

4.12 Entering the symbolic domain (Stage 1): choosing an entry point

• Purpose: Find where to start developing [source: MiM preview; 1999 interview].
• Steps: Let the client talk freely at first. Listen in a "not knowing" way for metaphors (explicit, or implicit such as "stuck" or "up against it"), gestures, and repeated words. Choose the entry with most energy or most connection to the outcome. Ask a developing or "like what?" question of it. Entry questions also include "And where are you drawn to?" (when the client looks at drawings of their landscape) and "And where are you going when you go there?" (when the gaze is fixed in one direction) [source: Contextually clean 2023].
• Example language: client says "I'm in a fog about work": "And a fog. And when you're in a fog, what kind of fog is that fog?" (illustrative).
• Watch for: the self-modelling cues in Section 2.

4.13 Developing a desired-outcome or resource landscape

• Purpose: Build the outcome (or a resource the client values) into a full metaphor landscape, so the client can experience it before problems are addressed [source: L&T 2011; Competency guidance 3c].
• Steps: Take the outcome words. For each symbol, establish its name, location and attributes before you move time or switch attention [source: Competency guidance 3c]. Keep adding: hold all relevant symbols in one perception ("make words physical") [source: L&T 2011]. In the authors' case, "at home with myself" became comfort at the core of the heart, flowing like a river [source: L&T 2011].
• Watch for: the client drifting back to problems (acknowledge, then go back to the outcome). A resource landscape that develops easily is a good sign.

4.14 Exploring effects

• Purpose: Test the outcome landscape against the client's life and earlier problems [source: L&T 2011].
• Steps: "And when [outcome], then what happens?"; "And when [outcome], what happens to [earlier problem symbol]?" Follow surprises (the authors' phrase is "follow the white rabbit"). If a new problem appears (e.g. the fish "frozen in mid-jump"), go back to PRO.
• Watch for: the problems that reappear are often the binding pattern. Move to 4.15 or 4.17.

4.15 Modelling symbolic patterns (Stage 3) and moving up a level

• Purpose: Find the structure that keeps the landscape the same [source: 1999 interview; L&T 2011].
• Steps: Use relationship and time questions to trace sequences. Notice when the same structure recurs across different metaphors and contexts. When single symbols stop shifting, ask the same questions of the whole pattern: recap the relationships in the client's words, then ask "And that's all like what?" [source: Rock and hard place; 1999 interview].
• Example language: "And the fish jumps, and freezes, and the river stops... and when all that, what would you like to have happen?" (illustrative recap using the authors' case symbols).
• Watch for: confirmation by repetition. The authors' example is a "going round in circles" metaphor that turns out to represent the client's whole life [source: 2017 psychotherapy interview].

4.16 Encouraging conditions for transformation (Stage 4)

• Purpose: Find the conditions under which the system would change by its own logic, and support them in the session. You do not cause change [source: 1999 interview].
• Steps:
  1. Ask symbols and agents what they would like to have happen.
  2. Ask what needs to happen for that, and whether it can (4.9).
  3. Look for and develop resource symbols, which may be subtle, anomalous or in the background [source: Rock and hard place].
  4. If the client doesn't shift naturally, ask about a resource symbol, the outcome, or a meta-perspective [source: L&T 2011].
  5. Then wait.
• Watch for: the signs of transformation in Section 2. Change can arrive unbidden [source: L&T 2011].

4.17 Working with binds and double binds

• Purpose: Handle self-maintaining patterns that seem to have no exit from inside [source: Rock and hard place].
• Recognise: (A) the client says the outcome "cannot or will not happen"; or (B) the client knows ways to reach it but evidence shows they are not reaching it. Common features are circularity, every option being unacceptable, oscillation, feeling trapped, a complex narrative, and self-reference. A double bind (Bateson) has a higher-level bind protecting the first, e.g. a belief that something terrible will follow if the problem is spoken about.
• Steps:
  1. Have the client self-model the pattern until they recognise it as a bind.
  2. Get a metaphor for the whole: "And that's [binding pattern] like what?"
  3. Ask for an outcome given the bind: "And when [whole pattern], what would you like to have happen?"
  4. Keep the bind as the context: "And when [binding logic], then what happens?"
  5. Turn concepts into perceivable form: "And how do you know [X]?"
  6. When the pattern happens live, ask "And what's happening now?"
  7. Stay put, wait, and bring the client back to the core of the bind when they wander.
  8. If it doesn't resolve, model the secondary bind at a higher level.
• Watch for: resolution is shown by a perspective that "transcends and includes" the incompatible sides. Keep a "let's see what happens" attitude until there is behavioural evidence over the following weeks. Avoid solving it for the client, sidestepping to an outcome, getting hooked on details, or switching technique on impulse.

4.18 Going live: the pattern in the room

• Purpose: Use in-the-moment shifts such as a pause, gaze change, tears or an unspoken reaction [source: Contextually clean 2023; Rock and hard place].
• Steps: When nonverbals change, ask "And what's happening now?" or "And what just happened?" Develop whatever is reported with the basic questions.
• Watch for: strong affect; see 5.3.

4.19 Maturing (Stage 5)

• Purpose: Help a change spread through and settle the landscape [source: L&T 2011; Rock and hard place].
• Steps:
  1. As soon as anything changes, develop the new state ("And what kind of [new]?", "And where is [new]?").
  2. Then ask "And when [new], what happens to [each other symbol]?"
  3. Check that every symbol's intentions are satisfied.
  4. Track how symbols and agents morph.
  5. Allow up to a third of the session for this [source: L&T 2011, on Grove].
• Example language: "And as the fish completes the jump, what happens to the river?" (illustrative).
• Watch for: doubts or a return of the problem. These are information, not failure; go back to PRO. Spreading, coherent change, such as the authors' case client feeling "comfortable in my own skin", shows it is working [source: L&T 2011].

4.20 Set down (closing)

• Purpose: End at a natural resting point [source: L&T 2011].
• Steps: When changes have settled and no new problems appear, ask "And is there anything else you need right now?" If no, ask "And is it OK to leave it here?" [source: L&T 2011]. Optionally invite the client to draw the final landscape [synthesis; drawing is a published tool, 4.22]. Don't summarise or interpret it for them [synthesis consistent with the principles].
• Watch for: a client still absorbed or tearful. Allow time to reorient (see 5.3).

4.21 Content-free trauma application (therapy use)

• Purpose: Work with trauma-related distress through the client's metaphors, without asking for a narrative. This was Grove's origin [source: Obituary]. The authors give two anecdotal examples: a survivor of childhood sexual abuse who worked only with a stone-room metaphor, and a soldier whose night sweats eased as his metaphor changed [source: 2017 psychotherapy interview]. The sources give no readiness criteria, dosing or trauma protocol [uncertain]. The safeguards below are [synthesis].
• Steps (safeguards are [synthesis]):
  1. Before any processing: assess for current safety (ongoing abuse, suicide risk), dissociative disorders, psychosis and substance misuse. Agree a stop signal. Make sure the client has grounding skills and support between sessions.
  2. Readiness criteria: the client can name an outcome; can develop a resource landscape (4.13) and stay oriented; can come back to the room when asked; and has stable daily functioning.
  3. Order: resource or outcome landscape first. Develop problem symbols only briefly and only in relation to the outcome (PRO).
  4. Dosing: keep problem-landscape work to short segments. Return to the resource symbol or the room before attention becomes absorbed in distress.
  5. Grounding and closure: reorient to the room, the present date and the body. Ask the client to place the landscape somewhere for now. Leave 10 minutes before the end.
  6. Memory caution: don't use source, age ("how old could…") or "what happens just before" questions to look for events. Treat any images as current experience, not evidence of past events. Never use the method to recover or confirm memories. The authors themselves "never asked" what happened [source: 2017 psychotherapy interview].
• Watch for: see 5.3. Use alongside an evidence-based trauma treatment plan where one is indicated [synthesis].

4.22 Drawing the metaphor landscape

• Purpose: Externalise the landscape. Useful when stuck, between sessions, or to choose an entry point [source: L&T 2011; Contextually clean 2023].
• Steps: Invite the client to draw their metaphors. Ask "And where are you drawn to?" as they look at the drawing, then develop that part with clean questions.
• Watch for: new symbols that show up only in the drawing.

4.23 Clean Space (Grove's late process, per Lawley and Manea 2017)

• Purpose: Use physical locations to let knowledge emerge from a network of perspectives. It suits issues that resist verbal modelling [source: Lawley & Manea 2017].
• Steps:
  1. Materials: paper, sticky notes and pens, with room to move.
  2. Ask the client to write or draw their outcome or topic and put it "where it needs to be". Then ask them to stand where they are now in relation to it.
  3. Ask "And what do you know here?" and whether there is anything else they know here about [topic]. Ask the client to name this space and mark it with a sticky note (Space 1).
  4. Ask the client to "find another space", then repeat: what do they know here, and what could this space be called. Keep each stay short (two or three questions). Build about six spaces.
  5. Linking: the client returns to each space in turn and is asked what they know there about the other spaces. Note repeated spatial metaphors and the shape of the network. The client may rearrange the notes.
  6. Return to Space 1 and ask what they know now, then "And knowing that, what difference does it make?" Continue until a change is reported.
  7. Close by inviting the client to collect their materials.
• Example language: "And find another space... and what do you know here?" (illustrative assembly of the published prompts).
• Watch for: if the client makes an emphatic statement or moves on their own, follow that rather than the routine. Stop using a space name that no longer fits. Silences of several minutes are normal. Typical length is about 40 minutes. Variants for couples and groups exist (Lawley & Way 2017, not drawn on) [uncertain].

4.24 Facilitator skill drills (for your own practice, not for clients)

The authors publish practice activities [source: practice activities page]. Only titles are drawn on here [uncertain on content]:

• full three-part syntax
• using your voice
• attending to every part of a statement
• flexibility in questions per minute
• directing attention to a kind of content
• using "when"
• extending vocal range
• nonverbal flexibility and stability
• adjusting in response to the client
• variations on the outcome question

A practical way to start is to tape a session, transcribe your questions, and mark any word that wasn't the client's or wasn't from a clean question [synthesis].

5. Handling problems

5.1 Stuck client (nothing shifts)

• Published responses [source: L&T 2011; 1999 interview; Rock and hard place]:
  • First check whether it is your stuckness, not the client's.
  • Then let the client set the direction: wait in silence; ask "And is there anything else?"; ask "And what would you like to have happen now?"; go back to the client's outcome in their exact words; or invite a drawing.
  • Ask about a resource symbol or a meta-perspective.
  • Move up a level, using "And that's all like what?" on the whole pattern.
  • Treat stuckness as a probable bind and model it (4.17). Don't switch technique.
• If you are entangled: stand up, create physical distance, and go back to the basic questions [source: 1999 interview].
• After 3–4 sessions without movement on the outcome: review the goal together, check for an unspoken higher-level bind (such as fear of what speaking would bring), and consider whether another treatment is indicated [synthesis].

5.2 "Resistance" and non-engagement with metaphor

• Authors' view: there is "nothing to 'resist'", because the therapist isn't trying to change the client. Self-resistance becomes part of the landscape to model [source: 2017 psychotherapy interview]. Example: "And when part of you doesn't want to, what kind of 'doesn't want to' is that?" (illustrative).
• If a line of questioning doesn't engage the client: "go somewhere else", i.e. switch to another word or symbol [source: 2017 psychotherapy interview].
• Clients who find metaphor odd or who think literally (some autistic clients, or clients early in treatment): explain briefly that you will use their own words and slow questions. Develop literal words with attribute, location and time questions. Don't force "like what?" or images. Offer drawing or Clean Space as non-verbal routes [synthesis]. The method has been applied in coaching with dyslexic adults (Doyle & McDowall, cited on Wikipedia; not drawn on) [uncertain].
• Clients irritated by repetition or the unusual voice: explain the rationale at the start. Shorten the syntax and speak more normally once the client is engaged [synthesis; flexible syntax is supported by the practice drills].

5.3 Overwhelm, flooding or dissociation

The sources say little beyond letting the client stop at any time and not pushing [source: 2017 psychotherapy interview]. Everything else in this section is [synthesis]:

• Early signs: fixed stare, loss of contact with your voice, rapid breathing, a frozen posture, or answers that drift into fragmented images.
• Response:
  1. Step out of the clean syntax. Speak in a normal voice and use the client's name.
  2. Orient them to the room, the date and their feet on the floor.
  3. Invite them to put the landscape "somewhere for now". You may use the client's own words for a container if they have offered one.
  4. Once settled, return to the resource landscape, or close.
• Afterwards: reduce how long you stay on problem material and go back to stabilisation. For clients with known dissociative disorders, work only within a phase-oriented trauma plan.
• Suicidal or self-harm content in metaphors (e.g. "I want to disappear"): step out of clean mode and carry out a standard direct risk assessment and safety planning, including lethal-means counselling where indicated. Clean neutrality does not override duty of care.

5.4 Non-adherence and between-session drift

• Symbolic Modelling sets no homework [source: none of the sources describe homework]. The authors advise wiping the slate clean each session [source: 1999 interview].
• Optional practices [synthesis]: the client draws their landscape between sessions and brings it in (develop it with "And where are you drawn to?"). They might also notice when the outcome symbol shows up in daily life. If the client doesn't do this, don't treat it as non-adherence. Start fresh with "And what would you like to have happen?"

5.5 Ruptures

• Common causes [synthesis]: the client feels the therapist is withholding, "robotic" or evasive because you don't share opinions or eye contact; or you have leaked your own content and the client senses being led.
• Responses [synthesis]:
  • Name the rupture in ordinary language and invite feedback.
  • Explain your reasons for the method's restraint.
  • Offer more ordinary conversation at the start and end of sessions.
  • Optionally, model the rupture cleanly with the client's agreement, e.g. "And when it feels like I'm not really here, what would you like to have happen?" (illustrative).
• The authors note that the rapport is with the client's landscape [source: 2017 psychotherapy interview]. Some clients still need visible relational warmth, so adjust [synthesis].

5.6 Method-specific difficulties

• Therapist leakage (adding your own words, metaphors or reframes): review recordings; go back to 4.1 [source: Less is More warns that leading or reframing pulls the client into the therapist's model].
• Jumping to fix a block: resolving a block too early can remove information, and the block may come back [source: Less is More]. Develop it first.
• Over-focusing on problems: use PRO to acknowledge them without dwelling on them [source: Competency guidance 4g].
• Bailing out of a bind: stay put (4.17) [source: Rock and hard place].
• Misusing specialised questions without their presupposition: see 4.10.
• Memory and suggestion risk: metaphor images, "child" symbols and source answers must not be read as evidence of real events. Don't confirm or explore them as history, and document that you are working with experience, not fact [synthesis].
• Scope: coaches should not work with severe mental disturbance [source: L&T 2011]. For clinicians, the method is a way of working inside the treatment plan, not a replacement for evidence-based protocols for specific disorders [synthesis]. Certification is assessed against published competency criteria by trainers and assessors. The sources do not name a single certifying body [source: Competency guidance] [uncertain].
• Evidence of lasting change: in-session transformation is not proof of change. Look for changes in behaviour over the following weeks [source: Rock and hard place]. Add routine outcome measures in clinical settings [synthesis].

Sources

Sources:

• Lawley & Tompkins (2011), Symbolic Modelling and Clean Language, chapter 4 in Innovations in NLP (author proofs; exact chapter title not verified) [cited as L&T 2011]
• Clean language and symbolic modelling in psychotherapy (interview, 2017) [cited as 2017 psychotherapy interview]
• Metaphors in Mind, Crown House look-inside preview [cited as MiM preview]
• Clean Language Revisited (Lawley & Tompkins, Rapport 2004)
• Moins est Plus / Less is More, French translation (Tompkins & Lawley) [cited as Less is More]
• Between a rock and a hard place (Lawley & Tompkins, on binds) [cited as Rock and hard place]
• Interview with Tompkins & Lawley, 1999 [cited as 1999 interview]
• Lawley & Manea (2017), Clean Space: a case study
• Lawley (2023), Contextually clean: specialised clean questions [cited as Contextually clean 2023]
• Guidance for competency criteria, Level 1 clean facilitator [cited as Competency guidance]
• Clean Language practice activities [cited as practice activities page]
• Obituary of David Grove 1950–2008 [cited as Obituary]
• Symbolic modeling, Wikipedia (secondary)

Books cited (not drawn on beyond the preview above):

• Lawley, J. & Tompkins, P. (2000). Metaphors in Mind: Transformation through Symbolic Modelling. London: The Developing Company Press.
• Grove, D. J. & Panzer, B. I. (1989). Resolving Traumatic Memories: Metaphors and Symbols in Psychotherapy. New York: Irvington.
• Sullivan, W. & Rees, J. (2008). Clean Language: Revealing Metaphors and Opening Minds. Crown House. (Cited in L&T 2011 for specialised questions; not drawn on.)
• Lawley, J. & Way, M. (2017). Insights in Space: How to Use Clean Space to Solve Problems, Generate Ideas and Spark Creativity. (Cited in Lawley & Manea 2017; not drawn on.)