The Structure of Change: A Response to Steve Andreas by Shawn Carson Sep 01, 2016

Shawn’s reply is in regular type; Steve’s comments are in italics.


 I have decided to divide my response into two parts; first to the two smaller items below which relate directly to the swish pattern. In a later post I’ll respond to Shawn’s discussion of the HNLP Meta Pattern, which goes far beyond the swish, to how we think about NLP processes in general.


I truly enjoy and appreciate intelligent discussion of NLP. Such discussions can make us all think more deeply about the principles that underlie this amazing discipline. I am therefore thrilled to have the opportunity to share ideas with Steve Andreas, one of the giants of early NLP development, regarding the Swish pattern.

Steve wrote a blog article regarding the Swish, I responded with my own article, and Steve commented in detail on my article. I guess this is now round 4! Steve made some great points in his last article, although on some matters we will have to agree to disagree. As Steve says, we will have to wait till there is serious scientific research into NLP to resolve these matters, something that seems unlikely in the near future!


To Read the Previous Exchanges Please Click the Links Below:

Steve’s Original Article 

Shawn’s Response

Steve’s Response to Shawn

Shawn’s Response to Steve


I would like to use this post to explain in detail the HNLP Meta Pattern, which Steve takes some objection to. But before I do that, I will make a few responses where Steve specifically requested.


Lifetime of a State

The first is the ‘life-time’ of a state (absent throwing logs on the fire). The estimated lifetime of around a minute first came to prominence (I believe) following the publication of Jill Bolte Taylor’s remarkable book, My Stroke of Insight. Dr. Taylor is a neuroanatomist and carried out her postdoctoral at Harvard Medical School. She talks about a state lasting 90 seconds (and who am I to argue?), which generated the famous ’90 second rule’ for state life-times. This is not a definitive rule as far as I can tell from consulting the Goddess Google, but the shortest time I could find anyone argue for is 6 seconds. Whatever the lifetime of a state is, it’s much longer than the time offered to a client to run the Swish, which casts doubt on Steve’s lap/butt joint theory.


As Shawn says, the question about how long a state lasts originated in his response to my writing that I thought that the “slingshot” swish, (in which the cue image first gets very small and distant, and then returns as the desired self-image getting bigger and closer) was less dependable because the two images only connect at a single point on the distant horizon, which I likened to a butt joint. In the standard swish, the cue image gets smaller and more distant as the desired self-image simultaneously gets larger and closer, so that there is a direction from cue to self-image at every point in the transition, what I called a lap joint. Shawn’s argument was that since states typically last longer than the swish transition, this wouldn’t make any difference—there would be ample overlap between states with both.


Steve uses a metaphor of anger-to-fear to argue that states can change quickly. Unfortunately anger and fear significantly overlap in terms of state; one of the major differences is blood-flow to the hands (higher in anger than in fear). Anyone who has been in a street fight knows that fight and flight are coded to allow smooth transition from one to the other; most street fights end when this anger-to-fear switch happens.


Anger-to-fear is an example, not a metaphor, and there are many others. Any situation in which an existing state is interrupted and quickly replaced with another was commonly used by Milton Erickson to change clients. Anyone who has experienced either surprise or embarrassment will have personal experience of a state shifting very quickly. Other states, often called moods, can last for a significant part of a day or more. Some, like depression or ennui, can last a lot longer, and are typically very hard to interrupt. So any generalization about how long a state lasts will be an overgeneralization unless the state is specified in greater detail.


Try less compatible states like ‘anger-to-forgiveness’ and you’ll feel the states battling it out over a longer period!


Actually, that’s not true at all. Our Forgiveness process—a specific application of Bandler’s “mapping across with submodalities”—requires some advance framing and information-gathering that takes some time, but the state shift in response to changing the submodalities of the representation—often a change in location is all that is needed—is almost instantaneous, and that is generally true of other applications of “mapping across,” for instance Resolving Shame, or Resolving Hate.

However, Shawn’s focus on how long a state lasts is irrelevant, because the submodalities transition in the swish doesn’t link states directly; it links two images. Since images can, and often do, change in a fraction of a second, a simultaneous transition will be much more dependable than the sequential slingshot swish. It took me a while to realize this simple, but crucial distinction (sometimes I’m a little slow, but I usually get there eventually). Thanks, Shawn, for raising this issue, and forcing me to think it through. 


Resource to End-State Pathway

Originally Shawn wrote:

“Now it’s important to know that if you as a coach, generate a big powerful resource state in your client (“awesomely confident”, say) and use that to collapse the trigger, change will happen. And, over time, that “awesomely confident” will transform into a lower energy (but more sustainable) ‘end state’.”

Steve replied:       

“I would be interested in Shawn’s understanding of how this transition from a big energy state to a lower energy state occurs. I would say that sometimes that happens. Other times the change will fall apart, and the person will revert back to the other side of the polarity.”


Shawn replied:

The second specific point Steve asked about is the idea of end state energy, and how that takes place. I have a fantastic reference experience involving a great friend Bella Rabinowitz (now passed), who had a fear of public speaking. She could only speak to a small group, and had to sit down to do so (because she then felt she was talking to friends). She worked through this with John Overdurf (the session was recorded and is offered for sale on John’s website, hence my freedom to discuss it) and Bella went through some big state changes. I saw Bella at an NLP practice group perhaps a month afterward, and she told me she had just spoken to 200 people at a prominent New York women’s group, and when I made some comment about how much she’d changed she made a characteristic gesture like she was brushing dust off her hands, her way of saying “no big deal”.


Shawn’s first response is to provide an example, a reference experience, which is not an answer to my “How?” question. Shawn’s next response is to provide a neurological explanation, and then one of his own.


So how does a problem become a powerful resource, then change to lower-energy ‘end state energy’? Well, to learn a new way of behaving you have to lock the new information into the hippocampus (which encodes most memories) long enough and powerfully enough for the new memories to form. This requires the release of dopamine to ‘lock’ the hippocampus and instruct it to begin forming long-term memories. Dopamine is typically released through strong emotionally-charged experiences. Hence (this is now my explanation) big resource states tend to release dopamine, lay down long-term learnings, and create change. However, once the change (the new state and new behavior) is wired into the brain, the brain seeks energy efficiency, by ‘dialing down’ the required state. That’s how Bella’s ‘big change’ became “no big deal” over the course of a couple of weeks.


I have great respect for neuroscience, and also for nuclear physics. However, I don’t think either one has much to tell us (so far) about learning and change.

Let’s make a clear distinction between memory formation and memory reprocessing. Memory formation is often facilitated by emotional arousal, but reprocessing the same memory often doesn’t require that. A lot of NLP consists of reprocessing memories and/or the conclusions based on them. A well-known example is the phobia cure, in which a terrifying memory is not changed, only the point of view is changed. In the phobia process, dissociation is certainly not “a big powerful resource state,” yet it is the perfect state for resolving a phobia.

In an article I wrote many years ago, ‘Selecting a Resource to Anchor,’ I pointed out that the appropriateness of a state is far more important than its intensity. A resource state for mathematics is mostly mental, so it’s not likely to be useful for skiing which is mostly physical, and vice-versa. Different skills require very different kinds of resource state.

When a client has experienced the phobia process, they immediately have the kind of blasé state that Shawn reports Bella having, with no need to “dial it down” so that it’s “no big deal over the course of a couple of weeks.” For a nice example of that, skip to near the end of my bee phobia video (8:34) when I ask the client to imagine a bee landing on her hand. When I ask her what that’s like, she replies matter-of-factly, “It’s like having one sitting on my hand.” When someone is very, very confident, it doesn’t look awesome at all, because it is “taken for granted.” Most of us are so totally confident that we can open a door that we don’t even think about it, we just do it.

Whenever someone wants to be “awesomely confident,” a thousand red flags flutter in my mind, as I wonder if they have the competence to go with it. Some people need confidence in order to demonstrate their competence; but far more often I see people who have too much confidence, without the competence to back it up. Years ago, I knew a guy who came back from a Tony Robbins workshop “awesomely confident,” and signed a year’s lease on 10,000 square feet of conference/office space; it remained largely empty until the lease was up.

It is fairly easy to elicit and anchor confidence, but developing competence takes a lot more persistent work and practice. People who are confidently incompetent are often a danger to themselves as well as others. As Robert Fulghum said, “Ignorance, power, and pride are a deadly mixture.”

Not too long ago I watched videos of a weekend workshop with an NLP trainer who was “awesomely confident.” He had over forty years of experience teaching NLP, and had written a couple of books. In one demonstration he “stacked anchors” for 5 different “awesomely powerful” states. When he attempted to elicit these states, the demonstration subject didn’t show any state changes, and at the end she reported no change in her problem state. The demonstration was a complete failure (which he didn’t acknowledge to the group) for two reasons: 1. He didn’t realize that some of the states were totally incompatible with others, and 2. His sensory acuity was so poor he didn’t realize that the demonstration subject wasn’t actually accessing any of the states, so he wasn’t able to anchor them—and he didn’t test his “anchors” so he had no way to check them.

To summarize, careful selection of a resource, to be sure it is appropriate for the problem state, is far more important than its intensity.