One of the people whose work I have enjoyed of late is Gadi Evron. I find that he and I approach problems and random things very similarly (although he blogs his results far, far more frequently than I do… mine just get saved up for classes, webinars and articles).

So, Gadi posted recently about his disappointment with NLP. It’s not the first time I’ve heard these arguments, and they all come down to a single, fundamental misunderstanding:

What we commonly call “NLP” is not science. Nor is it even scientific.

Most of this confusion comes out of the distinct issue that John Grinder called out in his book Whispering In the Wind. The thing that was originally “NLP” was a project that attempted to model successful people, notice the patterns of language and behavior, and replicate them. (This, Grinder refers to as “NLPmodelling“).

NLPmodelling was not scientific, but at least its principles were sound. Grinder and Bandler went and sat in the room with three strong therapists and learned to “act like” those therapists. They kept doing so until they were able to replicate the behavior. And then they continued to do so until they gained conscious ability to explain how they replicated the behavior.

While none of this was science, at least there was a principle behind it.

Where it all went to H-E-double-hockey-sticks is when they wrote down what they did and tried to explain how they replicated that behavior. This was a fool’s errand in some ways… there are grave epistemological concerns here – it’s beyond difficult to take your own behavior, translate it into conscious understanding and then try to convey it to others in language. It’s the same reason that great baseball players aren’t often good coaches – when you’re really good at something, it can often be difficult to teach others. Grinder once noted that when Bateson reviewed their work, his comment was: “Shoddy Epistemology.” Bateson was accurate, and this is where things started to get wonky.

This is because NLPmodelling is not what most people call “NLP”. When referring to NLP, most people are referring to the things that were written down – the hypothesis explanations that were posed by Grinder , Bandler and their colleagues/followers (e.g. Dilts, the Andreas’, etc.) to explain how they replicated behavior. These are what Grinder calls “NLPapplication“).

Unfortunately, because of the epistemological concerns, NLPapplication is about as scientific as me trying to predict the weather by sticking a wet finger in the air. Because we can hypothesize just about anything. I can observe how certain people act, and then make up any random example of why it must be true. For example, I could tell you that people are a certain way because of the position of the moon and the stars when they were born. How crazy would that be?

So, if NLP isn’t science, what are we to do?

Most people want to throw the baby out with the bath water. I’m a big fan of the original project – let’s look at people who get a particular result, and figure out how they do it.

But if you want to make it science, then turn around and figure out how it works.

Anyone who has looked at NLP has seen the following chart:

(Borrowed from http://completelymental.net/ )

The thing is, anybody who has tried to study whether it works finds that it doesn’t. Yet, many NLP people swear that there’s some efficacy in watching people’s eye patterns and using them to discern how people are thinking.

I was lucky enough to study NLP with Linda Ferguson and Chris Keeler at NLP Canada, and they get it. Linda was the first to point out to me that what Grinder & Bandler probably noticed (unconsciously) was the same set of patterns that Paul Ekman has noticed – we express many feelings and emotions in very small and quick ways with the musculature around our eyes.

So, while eye accessing cues don’t work, we find that paying close attention to that region of the face leads us to a detailed understanding of someone’s emotional state.

This is what happens when you approach a project without solid epistemology – you end up with many of the right behaviors, but the wrong reasons behind them.

And, sometimes, you end up with a whole pile of dogma and “true believers”. But that’s the subject of a different rant.

Until then, realize: NLP is not science. There is some useful background to take the tools and attempt to use them, and, even better, combine them with other, more useful science to figure out how to tie it together.

(As a shameless plug, I’m the one taking the lead on much of the “NLP-like” content at the SE Master Class. I say “NLP-like”, because it won’t be based on either NLPapplication or NLPmodelling. But anyone with an NLP background will find similarities on the things that really work in the real world, without much of the NLP and hypnosis dogma that goes around.)