Every field compresses its history into a founder myth, and NLP's myth is two men in Santa Cruz. Useful for posters, useless for accuracy. By the late seventies the raw material from Perls, Satir and Erickson had been turned into books, and the interesting question had changed. Not where did it come from, but who made it teachable at scale. The answer is a group of people whose names most practitioners never learn, and the field would not exist in its current shape without them.
Leslie Cameron-Bandler: precision and the dark side
Leslie Cameron-Bandler was married to Bandler for a time and was, by most accounts, the best pure modeller of the second generation. Her contributions were surgical. She pushed the Meta Model further and systematised what became known as the Well-Formed Outcome, the discipline of stating what you actually want in terms a brain can act on. Where Bandler's genius was improvisational, hers was architectural: she wanted the model to survive contact with ordinary practitioners.
She also forced the field's first honesty crisis. She said publicly that the standard NLP formats failed a substantial fraction of the time in her own practice, and set out to find why. The result was her work on what she called imperative self, and a stance that the field's claims should match its actual hit rate. A modeller who models her own failures. The field could use more of her.
Judith DeLozier: the ecological turn
Judith DeLozier was Grinder's collaborator and later co-developer of New Code NLP. Her influence moved the field's centre of gravity from techniques toward the practitioner's state. The question she kept pressing: what kind of person runs these patterns, and in what state do they run well? New Code's answer, change your state first or your technique will fail, is now so absorbed into NLP that people forget someone had to insist on it. She also carried Gregory Bateson's systems thinking into the field's bloodstream, which is where the ecological check comes from: no change is finished until you have asked what else it moves.
Robert Dilts: the cartographer
Dilts was the most prolific writer of the second generation and the one who organised NLP into teachable territory: strategies, the T.O.T.E. modelling formalism borrowed from cybernetics, and the belief-change work, including the classic integration formats. The work he is best known for is the set of Logical Levels: environment, behaviour, capability, belief, identity. Whatever a theorist thinks of its scientific standing, and it is a heuristic rather than a research finding, it gave practitioners a shared map for locating where a problem lives. He also co-founded the institute that kept long-form residential training alive. If the founders wrote the field's first drafts, Dilts wrote its reference textbooks.
David Gordon and the modelling formalism
David Gordon, another member of the original Santa Cruz circle, did the unglamorous work of turning modelling itself into a procedure a normal person can run: explicit steps from observation to description to transfer. His later work on modelling metaphor, with Graham Shepard, remains one of the field's more elegant corners. Every course that teaches modelling as a stepwise method rather than an aura is working from what Gordon codified.
What happened to the whole
Here is the honest part. There was no central committee and no standard syllabus, so the second generation also fragmented the field. Certifying bodies multiplied. Lineages diverged, each with its own emphases: Grinder's New Code, Bandler's Society of Neuro-Linguistic Programming, the Society of NLP versus the ABNLP versus a dozen others, Dilts's NLP University. A certificate means different things in each. This is not a scandal, it is just what happens when a field is built by strong personalities without a pope.
The practical consequence for a learner: you are never learning NLP. You are learning someone's NLP. Choose the lineage by what it can demonstrate, not by who blesses it.
Why this generation still matters
The patterns in the catalogue get the attention, but the second generation built the things that make training possible: outcome specification, state-first practice, ecological checking, the modelling procedure, and a shared map of levels. When we designed the Spirit of NLP programme, those are the load-bearing parts we kept.
For the origin generation, read two geniuses who could not explain themselves and the early history. For the man who is still building, see what Bandler is doing now. And for the map Dilts left behind, the levels you live on.