John Grinder (co-developer of NLP) emailed us in January 2006 to ask about David Grove’s work and our extension of it, Symbolic Modeling. John thought they might be:
“excellent examples of the agent of change discovering and coding strategies (in this case, verbal transactions) that are designed to provoke change without the imposition of the agent’s own material … and [that] respect the integrity of the client.”
He said he was writing an article focusing on ethics — the question of effectiveness would be handled separately — with the intention to include references and examples of how specifically it is possible to be dynamic and effective without crossing, what for him is an ethical line, by imposing on the client.
Recently, Inspiritive‘s Jules Collingwood posted a video on YouTube of John discussing just this issue in October 2008: What’s ethical in NLP training and counselling, and what’s not?
If the video does not play you can see it at: youtube.com/watch?v=3pFTMdq0v6Y
Some quotes from the video:
“Even Freud recognised over 100 years ago that to allow the imposition of the perceptions, the values, the beliefs of the analyst on to the client was a fundamentally unethical transaction. In particular because when a person is making a change they step into the unknown, they’re certainly vulnerable, they’re quite sensitive and suggestible, looking for a reference point in the unknown zone that they’ve stepped in to.
The relationship of rapport which I will presuppose is a requirement to be successful in this change-work indicates the most available reference point, in fact, is exactly the analyst or the agent of change or the context. So the analyst or the agent of change has a special responsibility to not make an imposition of their perceptions, beliefs or values. This is almost an impossible task. But it’s one worth struggling to achieve.”
“In training and in the application as an agent of change, I regard this as a very special responsibility. If you can’t make the distinction between process and content, go some place and learn it, but quit imposing these things [content], consciously or unconsciously, on your clients. You have absolutely no right to make this imposition. On the other hand, in order to know whether you’re doing this or not is itself a whole question about sensitizing yourself to the training and to the application of agent of change patterns.”
John’s video prompted us to publish our original reply to his enquiry below (written in January 2006, two years before David Grove’s death).
Our reply to John Grinder
What follows is our description of David Grove’s patterning. It is important to bear in mind a number of things. David is driven to keep developing his ideas. This means that any patterning we describe will sit somewhere in the sequence of the progression of his ideas. It is not necessarily a reflection of what he is doing today. Secondly, David has his own description of what he does that will almost certainly be different to ours, and will likely be couched in terms of his current interest.
One of the reasons we wrote our book, Metaphors in Mind: Transformation through Symbolic Modelling was to provide some stability to the field. Many people found it difficult to track David’s frequent developments and changing explanation of what he was doing. We noticed that the changes to his actual behaviour when working with clients went through a much slower trajectory than his explanations of what he was doing and why.
One of the fundamental differences between our description and David’s is that we rely more on a modelling and systemic philosophy, while David’s explanations (not his behaviour in our opinion), until recently, were based on a more linear, problem-solving approach. We call this kind of modelling ‘therapeutic modelling‘ to distinguish it from traditional NLP modelling — the process of acquiring and codifying excellence.
Historical Perspective
To give you a little background, there are three broad streams to the development of David Grove’s ideas:
In the 1980s he created a therapeutic process for working with client-generated metaphor. He devised Clean Language because he wanted to preserve clients’ descriptions as closely as he could without the “contamination” of his own constructs. He noticed that clients naturally used metaphor to describe traumatic events and painful experiences. And he realised that everyday questions did not encourage clients’ metaphors to remain in their awareness for long enough to have any effect (Grove said they have a “short half life” if not questioned properly).
In the 1990s he shifted to exploring the nature of inner space and time, and the utilisation of clients’ nonverbal cues. To do this he extended the set of Clean Language questions to encompass his “four realms or quadrants” of a client’s experience: I. Narrative; II. Form; III. Space; IV. Time [these are our labels].
In the 2000s his exploration has been around utilising external space and movement to affect therapeutic change through simple iterative algorithms. He initially distinguished this from his former work by calling it Clean Space and this lead him into the general field he called Emergent Knowledge.
[NOTE: A more detailed description the phases of David’s work and how these can be used in combination can be found at: Joining up the work of David Grove.]
We started modelling David Grove in the mid 1990s. We soon realised there were large areas of his early work that his students and clients raved about, but were not part of his current repertoire. We therefore used audio and video tapes of live sessions to model the first 10 years of his work. Our aim was to create a model that incorporated the range of his current and past approaches that would not be outdated by his future developments. In 2000 we published the results of our modelling in the book, Metaphors in Mind: Transformation through Symbolic Modelling.
Since then we have continued to track his latest innovations (they have appeared with alarming regularity!) and to produce generalised models which incorporate his new thinking. Five years on we can say our original model has demonstrated a certain robustness.1
Ethics of change-work
In relation to the topic of ethical practice, we think the fundamental contribution of David Grove is encompassed in the metaphor of clean. ‘Clean’ in the sense of ‘not contaminating’. We think ‘clean’ overlaps considerably with your expression “not imposing” and has a wider remit. ‘Clean’ is both a philosophy and a set of questions that are asked with clean gestures, voice qualities and line of sight.
The link between these the two levels – philosophic and behavioural – is the strategy level of selecting which question to ask of which aspect of the client’s verbal and nonverbal behaviour. Grove devised the questions, the philosophy of clean and is a master of strategy, but like so many experts, he sometimes struggles to explain how he does that. Our contribution was to codify how others could get similar results. Usually all three levels are included in the rubric, ‘Clean Language’.
The principal features of Clean Language include:
The use of a small set of questions that make maximal use of the client’s exact words and minimal use of words that are sourced in the facilitator’s own map (metaphors, assumptions and presuppositions).
The use of the facilitator’s gestures and lines of sight to preserve the spatial relations of the client’s inner perceptions from the client’s point of view. To do this the facilitator sets aside their own perceptual space and accepts the spatial relations inferred by the client’s metaphors.
The recognition of the metaphoric nature of internal perception and the value of clients attending to their autogenic metaphors (the field of Cognitive Linguistics, pioneered by Lakoff and Johnson, gives a theoretical basis for this approach).
A requirement to utilise only what is presented or presupposed by the client. In this respect the questions are, as Steve Andreas pointed out in Six Blind Elephants, entirely additive. They do not take away or reframe or seek to change the client’s experience.2 Instead they seek to acknowledge what is explicit and invite the client to quest for what is implicit. David Grove says he is interested in “what’s not there, but would need to be there, for what is there to make sense.”
In addition to the above, the principle features of Symbolic Modelling include:
A process whereby the facilitator bottom-up models the structure of a client’s metaphoric perceptions ‘in situ’ (i.e. within and around the client), thereby facilitating the client to self-model the patterns of their own map of the world.
An awareness of the embodied nature of cognition.
The facilitator, as Penny says, attends to what the client is attending to.3 And what or where the client is attending to is usually taken as the starting point for the facilitator’s next question.
The facilitator pays particular attention to the “psychoactivity” of the client’s perceptions – or as Michael Breen puts it, to “how the client responds to their own communication”.
Change occurs as a result of the feedback loop that is established between the unconsciously generated content of the client’s (metaphoric) perceptions and their conscious witnessing and exploring (self-modelling) of the nature of their experience – within a context of their intention to change for the better.
Our theoretical justification for this approach is that self-organising systems will naturally self-reorganise in a way that fits for the whole system – given the current structure of the system and its particular stage of its development.
Does 'clean' mean 'no influence'?
Some people mistakenly think ‘clean’ means ‘does not influence’, but this could not be further from the truth (what would be the point?). Clean Language is a highly influential and directive process, but not at the level of content. Clean Language’s influence is principally in it’s capacity to direct attention. However, since it restricts itself to use only the client’s precise content it is inherently respectful and “feels right” to them.
Content/Process distinction
Perhaps a word about the distinction between content and process would be useful here. All processes that involve language have to have content, and Clean Language is no exception. Clean Language uses the content of the client’s explicit metaphors, but crucially also the implicit metaphors that are embedded in their everyday language.
However Clean Language is not primarily a content model in the traditional sense. This is because (a) metaphor is a sideways shift from abstract concept in one domain to a more concrete analogy is another; and (b) while individual metaphors act as the components of a client’s “metaphor landscape”, the facilitator is modelling the relationships and patterns of the metaphors, and is attempting to work within the inherent organisation and logic of the client’s perceptions (while simultaneously conversing with the client at the metaphorical-content level).
When a client is working entirely in metaphor a facilitator will have no idea of the ‘actual’ circumstances of the client’s issue, and in that respect the process is ‘content free’.
David Grove’s latest work with Emeregent Knowldge aims to be ‘content free’ in another way. By using general algorythms that specify exactly what the facilitator says regardless of what the client says, the facilitator may know about the client’s content but there is no opportrunity for the facilitator to add in content.
To be more explicit, the role played by the facilitator is:
First, to “bless”, as David Grove says, whatever verbal and nonverbal behaviour the client presents and to facilitate them to take their own perceptions seriously.
Second, to facilitate the client to explore the nature, logic and ecology of those perceptions; to examine undervalued or under-visited areas; and to consider out-of-awareness relationships and effects.
Third, the facilitator is always on the look out for a spontaneous change in the metaphor landscape. Once a change occurs the facilitator invites the client to attend to that change and follow its consequences. This enables the client to discover whether a small change can spread into a larger, more significant change, or whether the system reacts to prevent the change from taking root.
Fourth, if the change peters out, the facilitator returns to facilitating the client to discover/model what just happened – because the system has spontaneously revealed an aspect of its nature.
Fifth, if the change spreads in a cascade-like effect, the facilitator’s role is to follow the change wherever it leads and when it comes to a natural resting point, and then to encourage its consolidation.
In addition, an experienced facilitator will quickly notice when the client’s description of a pattern in their life is happening isomorphically in the therapy session. Working in metaphor makes this easier to spot. At this point the facilitator ‘goes live’ and works with the moment-by-moment responses of the client to their own perceptions. This switch from the client talking ‘about’ the content of their perception, to them experiencing the process of their perceptions changing right here, right now, is often a key moment in the change process.
While all sorts of insight, changes of state and perspectives can occur during the session, some clients report that this process is a ‘gift that keeps on giving’ when they continue to notice unfolding consequences for many months and years after a session.
Remaining ethical
To extend the discussion outside of the frame of Grove’s patterning, any process can be misused. In order to remain ethical using Clean Language (or Symbolic Modelling) the facilitator needs, in our opinion, to:
Be keenly aware of the client’s desired outcome.
Have a ‘clean intention’ to support the client’s desire to change, but not in any predetermined way and without any notion of where they need to end up, or aiming towards ‘ideal’ states (e.g. “integrated”, “congruent”, “in their body”, “aligned” etc.)
Cleanly bottom-up model the client’s patterns of verbal and nonverbal behaviour.
Be really clear internally about the difference between their own stuff and that of the client.
The client’s desired outcome4 while being recognised as just one component of the entire client’s system should, for ethical reasons, be taken as a dynamic reference point within whose orbit the work remains.5
Because the client’s desired outcome changes during the process they rarely end up where they expected. However, by the facilitator maintaining a strong awareness of the client’s current desired outcome (which may be in entirely metaphorical terms) and using that to orientate the direction of their clean questions, the facilitator is deterred from ‘knowing what is best for the client’. This means the facilitator has to rely on their in-the-moment creativity that emerges from the interaction of their own system with that of the client’s and the local environment.
The facilitator starts from ‘not knowing’ and builds their model in-the-moment from the client’s description (this is bottom-up modelling as the facilitator is not expected to use neurological levels, perceptual positions, metaprogram distinctions, meta-model violations, or any other pre-given structures).6
The facilitator utilises what happens in the session and follows the directionality inherent within the organisation of the client’s system. When the session is finished, they end up ‘not knowing’ what the effects will be. Therefore the facilitator is wise to maintain a humble attitude, acknowledging that what they have participated in is but a small part of the client’s wider ecology (and, as Gregory Bateson was keen to point out, a part can never control the whole). As you may appreciate, this is quite a challenge for those ‘change agents’ with strong be-in-control, need-to-fix, need-to-understand, or get-it-right patterns.
Another ethical issue for us is the ability of the facilitator to pay attention to their own state – to appreciate and act on those internal signals which indicate they:7
Need to restrict their natural desire to ‘help’.
Need to restrict their everyday nonverbal behaviour which can unwittingly signal their approval/disapproval or personal interest/disinterest to the client.
Have detected a pattern.
Have recognised something is missing or inconsistent.
Feel drawn to follow a particular, often apparently inconsequential cue (clue?) embedded in the client’s verbal and nonverbal behaviour.
In summary, the facilitator needs to “trust the wisdom in the system.”
Penny Tompkins and James Lawley
28 January 2006 (with minor edits for this publication)
Notes
It is interesting to see some of our thinking about self-organising systems and emergence appearing in Grove’s latest descriptions. So who is modelling whom now?
With one exception, a question from Phase I of David’s work: “And would [X] be interested in going to [Y]?”
While practitioners of almost all therapeutic approaches will maintain that’s what they do, a close inspection of their verbal and nonverbal behaviour reveals this is rarely the case.
‘Desired outcome’ simply means a statement of how the client would like themselves or the world to be – it does not have to conform to the NLP Well-Formed Outcome conditions.
Of course, there are exceptions. For example client’s who are at a late stage of anorexia have often lost the ability to know what’s in their own best interest. However, it’s a principle of ours that the client has to prove by their behaviour that they are an exception to this rule before the facilitator may temporarily operate without a desired outcome statement from the client.
The previous acquisition of top-down processes and techniques is not wasted because they sit in the background of the facilitator’s awareness and provide templates which help guide the general direction of the facilitator’s questions.
Even though as you have pointed out, at that moment, they may not be able to put words to what they have noticed.
