Article Categories
[ Show ] All [ Hide ]
Clean Language
Article Selections
[ Show ] All [ Hide ]
Frequently Asked Questions

(answered by Penny Tompkins and James Lawley)

Who was David Grove?

What is Grovian Metaphor?

Why 'Clean' Language?

Where do Penny Tompkins and James Lawley fit in?

What is Symbolic Modelling?

What is NLP and how does it relate to Symbolic Modelling?

What is the difference between standard NLP modelling and modelling in a therapeutic context?

See also further questions and answers at Less FAQs.

Who is David Grove?

David J Grove, M.S. was a New Zealander of  Maori and European descent, whose unique psychotherapeutic approach, experience and style made him one of the world's most skillful and innovative therapists.

In the 1980s he developed clinical methods for resolving clients' traumatic memories, especially those related stemming from child abuse and war PTSD. He realised many clients naturally described their symptoms in metaphor, and found that when he enquired about these using their exact words, their perception of the trauma began to change. This led him to create The Philosophy and Principles of Clean Language, a way of asking questions of clients' metaphors which neither contaminate nor distort them. David Grove documented his approach in Resolving Traumatic Memories: Metaphors and Symbols in Psychotherapy (co-written with Bazil Panzer. Published by Irvington, New York, 1989).

David Grove continued to develop his approach and pioneered the clinical side of the 'healing the wounded child within' movement in America producing a number videos and audio tape sets (which are unfortunately no longer available).

During the 1990s his interests widened to include the examination of nonverbal behaviour, perceptual space and inter-generational trauma resulting in a therapeutic approach which integrates four domains of experience — semantic/cognitive, somatic, perceptual space, genealogical — and produces profound healing (see Problem Domains).

By the early 2000s, David developed Clean Space and Emergent Knowledge. He continued to innovate up to an including the day he died in January 2008.

David conducted seminars, workshops and healing retreats around the world. He was constantly developing new ideas and creative methods, so the articles and transcripts published on this web site should be considered as illustrative snapshots of what he was doing at the time, and not necessarily representative of the whole body of his work (see interviews with David and articles by David).

29 Dec 2000 (updated, 10 Feb 2005, Jan 2008)

What is Grovian Metaphor?

Metaphor is how we give meaning to the most important and complex aspects of our lives. As George Lakoff and Mark Johnson conclude, in their ear-opening book Metaphors We Live By: "Metaphors are not mere poetical or rhetorical embellishments ... [they] affect the ways in which we perceive, think and act . Reality itself is defined by metaphor."

The Metaphor Therapy developed by David Grove is a process that facilitates profound change by working within a person's own symbolic representation of their problem or issue. Client's words, gestures, sighs, lines of sight and other non-verbal cues provide entry to this out-of-awareness symbolic world. "Metaphor" David said "mediates the interface between the conscious and unconscious mind."

When a client says "I keep running up against a wall", David Grove not only assume this metaphor is an accurate description of the person's experience but also that it is the best and most complete description available to the client at that moment. Thus, what kind of wall it is, where it appears to be within the client's perceptual space, its size and shape, the direction of the running will all be symbolic of the 'replicating mechanisms' that "keep" this person repeating the particular behaviours they describe as "running up against a wall" — over and over again. The sum total of a client's autogenic metaphors is called their "metaphor landscape" which forms the context within which their symptoms are healed and they change. As the process unfolds new information becomes available to the client, enabling them to unstick stuck states, make new choices and change behaviours.

8 Jan 1998

Why 'Clean' Language?

Clean Language is at the heart of David Grove's Metaphor Therapy (and our Symbolic Modelling). At the core of Clean Language nine simple questions are used 80% of the time. (These are translated into a number of languages elsewhere on this web site.)

Clean Language is an extraordinary language because everything you, as facilitator, say and do is intimately related to what the client says and does. Since each Clean Language question takes as its point of departure the client's last verbal or nonverbal expression, there is minimal need for them to translate and interpret your words and behaviour. And because the client's response always informs your next question, the organisation of the client's information leads the interaction. Thus the entire focus of the process becomes an exploration of the client's model of the world from their perspective, within their perceptual time and space, and using their words.

Clean Language has three functions:

  • To acknowledge clients' experience exactly as they describe it.
  • To orientate clients' attention to an aspect of their perception.
  • To send them on a quest for self-knowledge.

Of course Clean Language influences and directs attention — all language does that. Clean Language does it 'cleanly' because it is sourced in the client's vocabulary, is consistent with the logic of their metaphors, and only introduces the universal metaphors of space, time and form.

June 2000.

See also: What constitutes Clean Language?.

Where do Penny Tompkins and James Lawley fit in?

We saw our first David Grove demonstration in 1993 and attended our first David Grove workshops and healing retreats in 1995. We were so impressed by what they saw and what we personally experienced that we wanted to know what David was doing that was so effective. Over the next four years we used a process called modelling to create a model of which way of working. This involved observing him work with clients (including our selves) and spending hour after hour poring over recordings and transcripts. We looked for patterns in the relationship between what he was doing and the way clients responded that contributed to the changes they experienced. These patterns were integrated into a generalised model which we tested and fine tuned — cycling through observation, pattern detection, model construction, testing and revision many times.

While our model is based on David Grove's work and incorporates many of his ideas, he has a different way of describing his approach. Our model was derived more from our observation of him in action than from his explanation of what he does. It was also shaped by our desire for others to learn the process easily and for it to apply to a range of contexts in addition to psychotherapy (such as education and business).

As well as employing many of David's ideas, we have also drawn upon cognitive linguistics, self-organising systems theory, evolutionary dynamics and Neuro-Linguistic Programming (NLP). The result, a process called Symbolic Modelling, is fully described in our book Metaphors in Mind: Transformation through Symbolic Modelling (2000) and demonstrated in the DVD A Strange and Strong Sensation.

There is more about us if you want it, and a list of our articles.

29 Dec 2000.

What is Symbolic Modelling?

In a nutshell, Symbolic Modelling is a method for facilitating individuals to become familiar with the symbolic domain of their experience so that they discover new ways of perceiving themselves and their world. It uses Clean Language to facilitate them to attend to their metaphoric expressions so that they create a model of their symbolic mind-body perceptions. This model exists as a living, breathing, four-dimensional world within and around them.

When clients explore this world and its inherent logic, their metaphors and way of being are honoured. They discover that their metaphors can limit and constrain or be a source of creativity and development. During the Symbolic Modelling process their metaphors begin to evolve. As this happens their everyday thinking, feeling and behaviour correspondingly change as well.

Some clients benefit just from having their metaphors developed with a few clean questions. For some the process leads to a reorganisation of their existing symbolic perceptions, while for others nothing short of a transformation of their entire landscape of metaphors will suffice. As a result clients report that they are more self-aware and at peace with themselves, that they have a more defined sense of their place in the world and how to enrich the lives of others.

The components of Symbolic Modelling — autogenic metaphor, modelling and Clean Language — can be used in three ways: to model successful strategies and states of excellence; to facilitate change; and to facilitate individuals and groups to create new metaphors (see diagram).

The components of Symbolic Modelling can be used together as a stand-alone process, or any one of them can be used in conjunction with other methodologies.

For a more extensive description, see the articles in the Symbolic Modelling category.

29 Dec 2000.

What is NLP and how does it relate to Symbolic Modelling?

The United Kingdom Council for Psychotherapy (UKCP) designates Neuro-Linguistic Psychotherapy (NLPt) as "Experiential Constructivist". This sums up how NLPt therapists work in a way that distinguishes them from other approaches. Of course there are lots of overlap, but differences define identity, so let's take those two words one at a time.


NLP facilitators consciously work with a client to change their "internal map" or "model of the world". Constructivists believe it is not possible to know anything about "the external territory" directly. All knowing is filtered through our map, and, as all maps are inaccurate representations of the world, it is the client's perception of the world, and not the world itself, that empowers or limits them. NLP is unique because it takes the map metaphor literally and because it assumes that people construct their map using five "internal senses" or "representational systems" (Visual, Auditory, Kinaesthetic, Olfactory and Gustatory). This map consists of representations of the Present State, the Desired State and Resources that enable the client to achieve their Desired State.

Rather than diagnose and categorise people or their symptoms from the outside, NLP therapists construct a model of the client's map from the client's perspective, and then seek to facilitate the client to change their map toward their Desired Outcome.

Why do NLP therapists work this way? Because of Richard Bandler and John Grinder's insight in the 1970's that experience has a structure and that when that structure changes so does the experience. (Actually, what they noted was that our representation of our experience has a structure and when that changes so does our experience. Practically it amounts to much the same thing.)

Each time we talk to a non-NLP based therapist about how they work we are reminded that this fundamental notion of NLP is still a radical idea. Some other approaches use some aspects of this methodology, but to our knowledge, none of them make it central to all that they do.


Although we construct unique maps of the world, we cannot construct any old map. The kind of body and neurology we have has evolved because we live in the kind of world that we do, and that massively constrains the kinds of maps we can create. Thus our maps are experiential in that they emerge out of our experience of the sensory world. Furthermore, unless my map of common human experiences is somewhat similar to yours we are not going to be able to communicate.

It works the other way round too. Our psychology affects our physiology. As Robert Dilts puts it, "Mind and body are one systemic process". By now most health professionals have caught on to the idea that the mind affects the body, but few understand it in such a direct way as it is meant in NLP.

Also, while we may join the client in their construct of past and future events, changing perceptual position, etc. we realise the client doesn't actually change the past or predetermine the future or step into someone else's shoes -- rather it is LIKE they do, i.e. these are metaphors. NLP is experiential in that we recognise we are working with the client's experience in this precise moment, even if they (and we) call it something else.

The Early Days of NLP

In the 1970's Bandler and Grinder studied some renown therapists and coded some of what they did in a number of models: Meta Model, Representational Systems, Milton Model, etc. These models were further applied to "the study of the structure of subjective experience" leading to yet more models and therapeutic techniques. Although NLP was originally conceived as the process of studying, coding and replicating (ie. modelling) excellence, NLP also came to stand for set of techniques that resulted from the modelling process.

Bandler and Grinder did not use NLP to model Fritz Perls, Virginia Satir and Milton Erickson because the notion simply did not exist at that time. And they never set out to define a stand-alone process, but  that is what they (and others) created.

The Early Days of Symbolic Modelling

In the 1990's we (Penny Tompkins and James Lawley) modelled a renown therapist, David Grove. As a result of our study we constructed a number of models of his work: the categorisation of Clean Language, the components and levels of embodied symbolic perception, etc. We discovered that David's primary approach was to facilitate his clients to self-model their own symbolic representations (NOTE: this is our explanation/metaphor, not his). These models were further applied to the study of the symbolic structure of experience leading to yet more models and an integrated therapeutic process, called Symbolic Modelling (SyM).

We did not use SyM to model David Grove because the notion simply did not exist at that time. In fact we used many standard NLP modelling techniques and we devised a few extra ourselves. We did not set out to codify a new method of modelling, it came about as a by-product of modelling David. SyM is still primarily conceived by us as a modelling methodology which has applications in psychotherapy, education, health, organisations, research, etc.

The explanatory part of SyM borrows ideas from cognitive linguistics, self-organising systems theory, evolutionary dynamics and NLP. As George Lakoff and Mark Johnson have shown, we not only continually speak in metaphor, we think, reason, understand and act in ways that are consistent with our metaphors. And that most of our metaphors are derived from the way human bodies interact with their environment, i.e. they are experiential through and through. Mind is an embodied phenomenon, from the electro-chemical level all the way up to the highest psychological levels.

SyM has made explicit several additional ways to model. SyM facilitates the client to self-model. To do this the facilitator/therapist also models the client's symbolic map (or 'metaphor landscape' as Grove called it) using the client's exact metaphors as the raw material for the modelling process. When used in a therapeutic context, the result is that clients often experience profound change.

Over the last 6 years, we have found that SyM is especially suited to working with 'higher levels' of experience – core beliefs, identity, sense of purpose, the spiritual – as well as complex and seemingly intractable issues, binds and double binds that are not amenable to traditional techniques.

In Conclusion

If what defines NLP is an experiential constructivist world view, the application of the study of the structure of subjective experience, the process of modelling and applying the results of modelling, then, in our opinion, Symbolic Modelling qualifies under every one of these criteria.

June 2001 (with slight revisions Jan 2004)

What is the difference between standard NLP modelling and modelling in a therapeutic context?

I think there is huge confusion about the difference between modelling in traditional NLP and modelling in Neuro-Linguistic Psychotherapy (NLPt) and Symbolic Modelling (SyM). Let me see if I can shine some light on this subject.

Classically in NLP, modelling is the process of a Modeller identifying an Exemplar (a person, or people who are exemplary of some behaviour of skill), constructing a model of how they do what they do, and facilitating other Acquirers to learn to take on the constructed Model (see Introducing Modelling to Organisations for more on the five-stages of a modelling project). John McWhirter calls this "Product Modelling" because the output is a physical product (representation) that if followed should produce a specific result. For example, the first five books of NLP were the product of John and Richard's modelling (NLP didn't even exist as a concept back then). Some, but by no means all, NLP Master Practitioner programmes still include a little on Product modelling in the form of "modelling project".

This is patently not what happens in NLPt, so why is modelling the foundation of NLPt? Because NLPt and SyM use a different kind of modelling.

(The NLPtCA Accreditation Procedures for an individual wanting to be registered with The United Kingdom Council for Psychotherapy (UKCP) say that applicants have to demonstrate knowledge and experience of "behavioural modelling" but no where is this defined.)

Modelling in a therapeutic or any facilitatory context for that matter, uses what Penny Tompkins and I call 'modelling in-the-moment' (or what Phil Swallow called "Modelling for the moment"). In this kind of modelling there is no 'product'. The therapist does not construct a formal model. Sure they may make notes but these do not constitute a consistent, coherent and complete model. The reason? The therapist does not have time to do this. Psychotherapy is a dynamic process. The client is always producing new information and the therapist has to constantly update his/her model of the client's model. Furthermore, the whole point of NLPt is that the client changes their model and very often that happens right then and there, in the session. As soon as this happens any formal model the therapist had constructed would be out of date.

The 'output' of modelling in-the-moment is the behaviour of the therapist. The therapist has to gather information, update their incomplete model, and respond using their model of the client model -- all within a few seconds! In NLPt and SyM, modelling does not produce a 'product' it results in a 'process' and more explicitly, a series of interactions which aim to enable the client to achieve (or at least move in the direction of) their desired outcome.

In Product Modelling there is an Exemplar and an Acquirer. Who plays those role in NLPt and SyM? I suggest that the client is playing both roles simultaneously -- and that's an ever-present conundrum of psychotherapy.

Every client is the Exemplar of getting the (unwanted) outcomes they so consistently get. Equally they want to Acquire their own desired outcome. And as an Acquirer they are faced with the situation that their current model works fine and seemingly doesn't have the room for a new and improved version. Furthermore, their existing structures have been honed, often over decades, to maintain themselves even when presented with repeated 'logical' solutions from family, friends, therapists, them self and other well-meaning helpers. Somehow the client's system has to figure out how to re-organise itself so that new possibilities become available when their system's natural tendency is to change only to stay the same.

Penny and I go even further, we think the client is also the Modeller in that they are self-modelling their own perceptions. (This, I hasten to add, is not a wide-spread view in NLP circles.)

A few other distinctions between NLP and NLPt modelling are:

  • Clients are exemplars of doing behaviours which they do not want and in contradistinction to Product Modelling a therapist has to learn how to not take on their client's strategies.
  • It is considered unethical in NLP for the Modeller to have an intention for the Exemplar's model to change, while psychotherapists are paid to have just such an intention.
  • I really want to highlight the following: Who owns the outcome? In Product modelling the Modeller decides what they want to model (their desired outcome) and the Exemplar agrees to be part of the process but rarely has any outcome beyond being helpful. When modelling in-the-moment, the client owns the outcome and the NLPt therapist agrees to set aside their own agenda and work toward the client's outcome.
  • The situation gets more complicated when the therapist runs an NLP technique which has a pre-given outcome e.g. a 'Forgiveness Pattern', which the client may not have asked for – maybe because they did not know it was possible. (I am not commenting on the efficacy of the technique, I pointing out that who owns the outcome can get blurred once the therapist moves out of a bottom-up modelling in-the-moment frame and into a top-down technique frame.)

Finally, modelling is not the province of NLPt alone. Many of the founders of the major schools were brilliant modellers. In addition to Satir, Erickson and Perls (who were the original exemplars for NLP), I'm thinking of Freud, Jung, Berne, Lang, etc. The two key differences with NLPt is that:

(a) the followers of those great minds were expected to apply or improve the models created by the founders. They were not explicitly tasked with continuing to model (although all of the best therapists I have observed have developed their own semi-conscious ways to model); and

(b) John Grinder and Richard Bandler gave the world a way to conceptualise working directly with the organisation of the client's map, i.e. the structure of their subjective experience.

23 October 2005

For more on the different type of modelling see What is Therapeutic Modelling?

13-16 May 2022

W. Kirby, England

*** in person ***

Six of the world's leading clean facilitators will support your personal development using the whole range of Clean Language approaches

And, you can observe them working with others and ask questions to find out how they do what they so well.

view all featured events