First published in Rapport, Issue 21, Autumn 2010
What Makes a Modeller?
discusses modelling with experts
James Lawley and Penny Tompkins
Modelling is at the
heart of NLP: it’s the methodology that leaves behind the field’s
famous “trail of techniques”. But for many students of NLP,
challenged to undertake a modelling project as part of their Master
Practitioner course, it can be something of a puzzle.
I’m sure I’m not
alone in struggling to get started, overwhelmed by competing
methodologies on the one hand, and baffled by the missing links in
trainers’ stories of Bandler, Grinder and the “wild days” on
What actually makes
a great modeller? Penny Tompkins and James Lawley should know. They
are not only expert modellers themselves, but have undertaken a
number of modelling projects in which they have modelled great
They have studied
with many of NLP’s best-known modellers – including John Grinder,
Robert Dilts, John McWhirter, David Gordon and Graham Dawes – and
have been involved in a huge range of modelling projects. Modelling
has become a habit for them, to the extent that Penny said: “We
almost can’t not model nowadays”.
In their first major
modelling project, the couple modelled psychotherapist David Grove,
creator of Clean Language. What began as a one-year commitment
extended to four years before they eventually published their book,
Metaphors in Mind, in 2000.
moment came when they realised that Grove wasn’t trying to change
his clients – he was modelling their inner landscapes – a process
Penny and James called “symbolic modelling”.
psychotherapists themselves, they model their own clients’ inner
worlds and help them to discover more about themselves, how they do
what they do, and how they can have more choice – to “self-model”.
As trainers, they model their students’ learning processes.
modelled well-known NLP modeller Robert Dilts over the course of a
weekend, as part of an event organised by Fran Burgess and Derek
Jackson of the Northern School of NLP. The results of that project
are now available in a comprehensive report – including video clips
and transcripts - on their website.
What is Modelling?
NLP has very little to do with either Naomi Campbell, or building
miniature plastic aeroplanes! It’s more akin to scientific
modelling: Wikipedia describes this as “the process of generating a
model as a conceptual representation of some phenomenon”.
But there are wheels
within wheels. In NLP, modelling is typically viewed as a process
whereby a modeller:
exemplar (a person, or people who exemplify some desired behaviour
information about what the exemplar does;
model of how they do that;
using the model gets similar results to the exemplar.
- goes on to use
the model themselves, or facilitates others (acquirers) to learn how
to apply the model.
There are lots of
ways of doing each stage.
Penny and James
distinguish between this process, “product modelling”, and the
related process of “therapeutic modelling”. In the latter, a
therapist constructs a working, in-the-moment model of their client’s
“model of the world”, which they use to guide their
interventions. This may be held more or less consciously by the
therapist – there is no need to formalise it, to write it down, or
to share it.
For a number of
articles about modelling including “How to do a modelling project”
When Penny and James
were modelling David Grove, there were multiple levels of modelling
The client was
learning more about themselves – self-modelling
learning about his client’s inner world – therapeutic modelling
Penny and James
were creating a model of Grove’s process with a view to making his
skill available to others – product modelling.
It seems that
they’ve come a long way. As they explained, when they began their
work with David Grove, they realised that most of what was known
about NLP modelling was implicit. John Grinder and Richard Bandler
had written up the results of their early modelling, but not how they
did it. So Penny and James had to “reverse engineer” how the
founders of NLP had done it before applying that learning to their
modelling of Grove.
To make things
worse, Grove was a very reluctant exemplar! He initially agreed to be
modelled only on condition that he didn’t have to answer any
questions and that they didn’t mimic what he was doing. They could
attend his seminars, but only as ordinary participants.
Gradually, he became
curious about what they were doing and became a close friend. But he
remained reluctant to answer questions about what he was doing. When
he eventually agreed to be interviewed, it had to be in a hot tub,
with the recording equipment well out of sight. James said: “So
much of him was out there running around in the client’s landscapes
that he didn’t spend time self-reflecting on his own internal
In contrast, Robert
Dilts was an enthusiastic subject, who loved to talk about his
modelling process – as befits the author of Modelling
with NLP. By the time Penny and James
modelled him, they had had plenty of practice and knew better than to
try to model his entire process in a few hours. Instead they selected
a small part to pay attention to – how he selects “what is
essential” while modelling.
It’s a superb NLP
modelling pedigree. If anyone can help those struggling Master Prac
students, it’s Penny and James.
I should declare an
interest. I’ve been an avid student of Penny and James’s work for
several years, and would happily interview them about modelling for
hours. In doing so, I’m a modeller, too.
And in this article,
we have two pages – and that brings me up against the modeller’s
dilemma. I have many pages of notes, two hours of audio recording.
What do I select, from all the information I have, to include in my
“model”. What’s essential? And how I can I best present the
model so that you, the reader, are able to use it?
As James observed,
this part of the modelling process can feel almost violent: after
making pristine observations of your exemplar’s words and actions,
you now find yourself changing them to make them easier for others to
newly-created model is born of the modeller’s
map: it draws on the modeller’s
knowledge and experience, and is ultimately limited by their
imagination and other mental capacities.
So, I’ve selected
one key piece, which I think you’ll find interesting because it’s
new, and because it provides something which many NLPer will find
useful and relevant. It’s a new model, which Penny and James have
not published before. And it encapsulates Penny and James’ learning
from various modelling projects over the years.
It’s a list of the
core skills required of a good modeller.
orientation, having a strong sense of
what your purpose is in conducting a particular modelling project and
being able to maintain that while navigating the unfamiliar and
often confusing territory that is the exemplar’s world. The
outcome remains a “dynamic reference point” that guides the
modeller throughout the project.
comfortable with large amounts of
information, and with not knowing how it all fits together. Penny
and James’s experience suggests that most, if not all, modellers
will be inundated with information, much of which will not be
relevant, and they need to have some way of coping with it all.
detection and split attention. A key
NLP distinction is between content and process – between what a person thinks and/or does, and how they think and/or do it. An expert modeller is able to pay just
enough attention to the content to keep the process moving, while
noticing and investigating patterns at the process level. Penny
said: “You can know there’s a pattern before you can articulate
This is something of
a work in progress: there may be crucial pieces missing, or it may
not be as elegantly simple as it’s possible to make it.
The proof of its
value will be in the results it gets. If you’re an NLP modeller,
why not try developing these skills – and let me know what happens