The Room in Duluth, 1997
In 1997 I took days out of my second year at university, flew from Singapore to Minnesota, and walked into a training room at the Black Bear Hotel and Casino in Duluth to learn this work from Richard Bandler. Two friends came with me: Devon, who has since passed on, and Wiki Tay, who still trains and speaks.
The program ran close to four thousand US dollars before flights and accommodation. My father funded it, and he asked for nothing back. I have carried the weight of that gift for close to thirty years.
What stays with me is the texture of the room. Nobody handed us a script and asked us to memorise it. We experimented on each other, changed protocols mid-run, and reported what happened. Bandler installed fun, energy, and a particular kind of audacity, and the audacity had a structure underneath it. He wanted to know what the specific person in front of him was doing inside their own head, and he treated that answer as more interesting than any technique in the manual.
I flew home and ran my first program with seven participants. Two of them were my parents. When I asked my mother what she had taken from it, she said state management, that she could manage her emotions better, and that it gave her a way to become a better version of herself. My father said he finally had a framework and a language for things he had been thinking for years.
Both answers described something that had happened inside them. Neither described a certificate. I have been thinking about that gap ever since.
What He Said the Work Was
Bandler has spent decades describing this discipline as an attitude rather than a collection of techniques, and the point keeps getting lost in translation. In a long-form interview released in June 2025, he traced the origins of the work through modelling and hypnosis and kept returning to the same frame: curiosity about subjective experience as the starting condition.
The clearest written version of that framing comes from Joseph O'Connor and John Seymour, who laid the discipline out in Introducing NLP in 1990 as three layers. Attitude sits at the base. Methodology sits in the middle. Technique sits on top. Training markets the top layer, because a technique is visible, demonstrable, and testable on the last afternoon of a course.
The attitude underneath has a specific shape. It is the stance that another human being is doing something systematic inside their own head, that the structure of it can be found, and that the finding carries more value than the applying.
Read the early books and the stance shows on every page. The Structure of Magic in 1975 and Frogs into Princes in 1979 are records of two men asking how unusually effective therapists did what they did. The techniques arrived afterwards, as the by-product of a question. These days the technique arrives first and the question rarely comes at all.
Here is the practical cost. A practitioner who opens with curiosity notices what this person is doing. A practitioner who opens with a technique notices whether the technique landed. One of those practitioners can work with anybody. The other works with the people who happen to fit the move.
Why the Certificate Became the Product
The training industry has an accounting problem, and the problem has a natural solution. Curiosity is invisible on a resume. A certificate is visible, portable, and verifiable. The market builds the certificate, prices the curiosity into it, and the customer receives the part that photographs well.
Bandler and Grinder had a hand in this. They certified the first generations of practitioners, and the Society of NLP that grew from that work set standards for trainers and for assessment. The structure was reasonable for its time. Three decades later the same market contains a weekend seminar and a full practitioner pathway that both produce a certificate, and the buyer has no straightforward way to tell which one changed their capability.
I hold a Master Trainer certification, and I have trained many people toward theirs. The certificate carries real value. It sets a floor for assessment, it gives a trainer something concrete to test against, and it tells a client that a practitioner submitted to review by somebody other than themselves.
The certificate has a limit as well. It records that you were in the room and passed the assessment on the day. It does not record whether you still ask what this person is actually doing.
That limit shows up commercially, which is why I raise it with founders rather than with trainers. A coach, a consultant, or a sales leader sells judgement. A client can feel the difference between a person reading their situation and a person running a routine, and the second one gets replaced as soon as a cheaper routine appears.
A Warning About the Moves
In Orlando earlier this year, I sat in a session with John Lavalle, president of the Society of NLP. He said something that reframed the whole argument for me. Techniques are seductive, he said, because they make practitioners believe problems can be solved with techniques alone. The truth, he said, is that techniques were invented because of the client, and there are many pathways to help a client.
Read the second sentence twice. Each technique is a downstream artifact. Every one of them came from a specific human being with a specific structure, and the toolkit grew because no single pathway fits every person who walks in.
What fails when the attitude is missing is predictable, and I have watched the same three failure modes in training rooms for years. A practitioner with no foundation applies moves without state work, and the moves read as theatre. A practitioner with no flexibility carries one move per problem and hits a wall with the second client. A practitioner with no structural understanding copies what I do on stage, delivery included, and cannot explain why any of it worked.
The last one costs the most. When you cannot explain why a move worked, you cannot adapt it. You perform a ritual and hope the ritual travels.
The Loop That Makes Curiosity Operational
Curiosity sounds soft until you give it a mechanism, and the mechanism already has a name. George Miller, Eugene Galanter, and Karl Pribram described the test-operate-test-exit unit in Plans and the Structure of Behavior in 1960: test the current state, operate, test again, exit when the test passes. Bandler expanded that loop into a fuller architecture inside Design Human Engineering, with sensory criteria at each test, a set of available operations, and an explicit decision point choosing between exit, loop, switch, and change of criteria.
Strip the vocabulary and the architecture reads as an attitude in motion. You test before you act, which forces you to find out what the person's current state is rather than assuming it. You keep more than one operation available, which forces flexibility. You test after, which forces you to look at results instead of intentions. You hold a decision point, which forces you to treat a failed test as information rather than a verdict.
That loop is the difference between a practitioner and a technician. A technician runs the operation and reports that the protocol was completed. A practitioner runs the test, reads the result, and decides what happens next.
Supervised repetition is where this turns into skill. K. Anders Ericsson and his colleagues published their work on deliberate practice in Psychological Review in 1993, and the finding that matters here is simple to state: expert performance comes from repetition with immediate, accurate, corrective feedback. Repetition alone produces fluency. Repetition with a trained eye watching produces competence.
Copying a Move and Modelling a Structure
The word modelling gets used loosely, and the loose use hides the whole method. Copying produces a person who can reproduce a demonstration. Modelling produces a person who can rebuild the capability in somebody else.
The method Bandler and Grinder applied has three moves, and Robert Dilts documented them thoroughly in Modeling with NLP in 1998. First, elicit the structure: find out through questioning and observation how the person sequences their experience, which representations they use, and what triggers what. Second, contrast: compare somebody who can do the thing with somebody who cannot, because the difference carries the model. Third, encode and test: write the pattern down in a form another person can follow, then teach it and measure whether the capability transfers.
The contrast step is the one people skip, and skipping it produces the expensive failure mode from earlier. Everybody in a training room can copy the demo. Finding the difference between the two performers requires you to hold two structures in your head at once and to keep asking questions after you think you already know.
For a founder, the same method answers a commercial question. Your top salesperson outperforms the rest of the team by a wide margin, and you want to scale that. Copying the script produces four people who sound like your top performer and close like nobody. Modelling the structure produces a hiring profile, a training sequence, and a set of tests you can run on a Tuesday afternoon.
The Woman on Stage in Orlando
One case from Orlando stays with me. On the first morning, a woman came to the stage carrying a heavy internal load: looping negative imagery and a set of voices to match. Bandler's work with her looked unremarkable from the audience. No dramatic induction. He used her own conscious mind as the vehicle, embedding possibilities beneath the content of the conversation, and put her resistant thinking into a loop that produced something useful rather than something corrosive.
On stage, the shift was barely visible. (Anyone expecting fireworks from this work has been watching the wrong room.) Over the following days it compounded. Heavy energy turned into optimism. The negative loops gave way to a flexible response to whatever came at her. By the end of the week she was treating criticism as information instead of an attack, and she was doing it without instruction.
Notice what the case demonstrates. He was building infrastructure so that future difficulty would meet a nervous system that could flex. That target required reading her particular structure, and no standard protocol supplies the read. The read is the attitude made concrete.