I create my model of the client’s model, their metaphor landscape (ML) within and around them, i.e. I put each of their symbols in the same place they do, from their perspective.

When I am asking a question of a symbol or area of a person ML I will often address it by ‘pointing’ at it’s location with my eyes, head, hands.  When I do that, I cannot maintain eye contact – and I don’t want to. I don’t want the client to attend to my eyes but rather to their landscape. (I had an interesting experience once when a client’s landscape was “deep within your (i.e. my) eyes”, but that’s another story.) e.g. When referencing a demonstration client’s metaphor of a light switch, I looked up and to just below the ceiling – because that’s where he had told me it was. This saves the client having to convert my nonverbals back into their perspective as in traditional body-mirroring rapport. When the client’s symbol was in his head I looked at his head and not in his eyes.

I prefer to look at the client when they are not looking at me because that means they are engaged with their inner world and not me. Then I can observe their nonverbal cues, especially when they are ready for the next question.

All of this is in service of what David Grove suggested, that “the I-ness of the therapist should appear to cease to exist.” You’ll notice that apart from the ‘set up’ question “Where would you like me to to be?”, clean language questions never refer to the facilitator’s ‘I’ or ‘me’ e.g. “Tell me some more about …” or “I was wondering ….”.

Now to the second part of your question am I visualising a spatial model? Well yes and no. In NLP-speak, relatively, Vi (imagined visual) is my weakest representation system, so I rarely ‘see internal images’ (to use a metaphor!). Instead I have a vague visual sense of a symbol and a strong map of the spatial arrangement of their landscape – much of which is coded proprioceptively.

Once I have a symbol’s name and address (attribute by which the client refers to it, and it’s location relative to the client’s physical position), I am mostly interested in the symbol’s function within the system, i.e. how does it contribute the the maintenance of the system? Or, how will it contribute if it is part of a desired outcome landscape.

Also, in the pauses, I am ‘surveying the landscape’. My attention is visiting the symbols in their location to (a) remind myself where they are and their characteristics, and (b) to muse and wonder about their place and function in the landscape. During this time I will not be maintaining much eye contact.

Now it gets interesting. Systemically speaking ‘function’ is always relational. And, following Bateson, most relationships ‘between’ symbols do not have a location and cannot be seen, heard or touched. e.g. If two symbols are ‘in conflict’, you can describe the behaviours of each symbol when conflicting but you cannot easily represent ‘the conflict’ directly (although you can say what it is like!). You can represent the word ‘conflict’ but think of how much you know about conflicts that is not contained in the word, where is it?  It is, in my language, non-representational or background knowledge. As Bandler has said: it’s what happens in the ‘arrow’ (of a say a V -> K) coding that matters.

Some people (inside and outside NLP) deny the existence of knowledge that cannot be represented in VAK but I challemge them to represent the difference between black and white (without using black and white or providing a list of the attributes of each, which is not the same a describing the difference between them).

So, if relationships and patterns are rarely able to be represented, and given I reckon they are the most important part of a model, how do we code them while modelling?

My answer is to code them in embodied metaphor. i.e. my mind-body can ‘do’ or ‘hold’ a map or schema of metaphors for the relationships and I can remember the client’s labels so I can refer to them with words (or gestures if needs be).

When I am replaying to myself the way their system is organised I am in my landscape and will unlikely to looking deep and meaningfully into their eyes at that time.