Neurolinguistic Programming: A Linguist Druid’s Review

As I described in this previous post, one of the requirements of the Magic Spiral in the candidate year in the AODA is to learn about magic through reading and meditation. The books I selected to start with were three on “neurolinguistic programming” by Richard Bandler and John Grinder. I started with Bandler’s book, Use Your Brain for a Change, which is an edited set of lectures from the 1980s, and The Structure of Magic I & II, which were written in the 1970s. Use Your Brain for a Change especially comes highly recommended. As a linguist, I was very interested to see how linguistics would play into these techniques. I’ll lay out some of my thoughts below.

What is NLP?

barackobamareadingNeurolinguistic programming is claimed to be an atheoretical set of axioms and techniques for personal development and therapy. (Is it really atheoretical? I’ll talk about that below.) It is extremely popular, and has been for many years. However, there are a lot of questions about its real effectiveness and its claims to be “scientific”; if you want to read a real downer NLP-smackdown, check out the Wikipedia article. I’m not going to talk about whether it’s “scientific” here, but whether it’s coherent, whether it makes sense, and whether it’s useful.

Will the Real NLP Please Stand Up?

First of all, Bandler’s 1980s book is vastly different from the other two. In Use Your Brain for a Change, Bandler explicitly states that neurolinguistic programming is not a “theory”, but a toolbox. He then goes on to present a set of techniques with not much explicit theoretical underpinning.

But there’s just no such thing as a toolbox without any theory behind it! The theory is there, whether it’s explicitly stated or not. This is even more obvious in the other two books, where, again, the authors state that they’re not interested in propounding a new theory of therapy — and then they go on to propound a new theory of therapy. Interestingly, the implicit theory behind Use Your Brain for a Change is different from the theory behind The Structure of Magic! Calling them both “neurolinguistic programming” is just weird. I’ll explain what I mean below.

Use Your Brain for a Change: No Neurons, No Linguistics, Just Programming

Use Your Brain for a Change has no neurons and almost no linguistics, but it does have programming. The theory hidden in the book seems to be that all mental states — emotions, thoughts, beliefs, insanity, etc. — are associated with and triggered by sensory input; and by changing the associations and triggers, you can change your mental states. For example, Bandler claims that you can change your mental state from “confusion” to “understanding” by (1) finding out what sensory inputs you associate with the confusion, (2) finding out what sensory inputs you associate with something you understand, and (3) training your mind (through repetition — this is the programming) to associate understanding-sensory-inputs with the confusing topic. Now, in the book, Bandler says he has seen this work many times. It didn’t work for me. As far as I could tell, my understanding of, say, linguistics is not associated with any particular sensory input. I just know it. I could not use any sensory inputs of associated with linguistics to help me come to a better understanding of quantum mechanics. The first axoim of Bandler’s “theory” — that mental states are always associated with sensory input — just seems to be false.

That’s not to say that neurolinguistic programming would never work. It might depend on whether or not the mental state you’re trying to change is associated with some kind of sensory input. For people who are very visual thinkers, visualization techniques, I imagine, would be extremely effective. Repetition and association are without doubt extremely good ways of training neural nets like the human brain. That’s how magic works, too (fancy that!).

The Structure of Magic: No Neurons, No Programming, Just Linguistics

The Structure of Magic is entirely different. It has no neurons and no programming, but a huge wallop of linguistics. The authors state again and again that they’re not offering a new theory of therapy, but a “new way of talking about” therapy. But of course that’s just what a theory is — a way of talking about something.

Bandler and Grinder take the major elements of 1960s linguistic theory and glom them wholesale onto therapy. They claim that the structure of language, as discovered by linguists such as Noam Chomsky, can serve as a model for the whole mind. They claim this without any attempt to provide supporting evidence. They claim this even though almost no linguists or brain scientists believe it, then or now — least of all Noam Chomsky. For his entire career, Chomsky has been a bitter defender of the idea that language is a separate, unique, and fundamentally different kind of mental function. Chomsky may well be wrong about this, and personally I think he is — but nowhere do the authors indicate that their position is not only controversial, but rejected by Chomskyan researchers.

Shotgun Marriage of Linguistics and Therapy

The uneasy marriage of linguistics and therapy has begotten two unfortunate offspring.

Outdated Linguistics

First, Chomskyan linguistics is a young science — even younger than psychotherapy — and it has changed drastically over the last 50 years. Nearly all linguists, including Noam Chomsky, have abandoned the ideas of “Deep Structure” and “Surface Structure” that form the underpinning of Bandler and Grinder’s theory. Chomsky now advocates a hypothesis in which even simple sentences are described by perhaps a dozen levels of representation, not just two. Other linguists keep the same basic two levels of representation, but postulate dozens of (or even infinite!) candidate “surface structures” that are evaluated in parallel as the sentence is spoken. It’s hard to see how these new theories of linguistics, which do a much better job of explaining language, can be made to match up with the theory of therapy outlined in these books.

The Map is Not the Territory…In Fact, it’s a Map of the Wrong Territory

Second, and even more seriously, the mapping from therapy to linguistics just doesn’t work. Bandler and Grinder take concepts from linguistics and try apply them to therapy, but the mapping is quite faulty. Let me be specific.

In Chomskyan linguistics of the 1960’s, when someone gets ready to make a sentence, the language faculty first generates a “Deep Structure”, which is an abstract representation of the meaning they’re trying to convey. Then, a sequence of transformations in the speaker’s grammar change the Deep Structure into a Surface Structure, which is an abstract representation of the actual sentence as it will emerge. Then the speaker’s grammar flattens this Surface Structure into a string of sounds, and out comes a sentence.

Here’s an example. A speaker wishes to convey the idea that she was abducted by aliens. She forms a Deep Structure which represents the idea. The Deep Structure contains the information that “aliens” did something, and that something was “abducted”, that the abduction took place at some prior to the act of speaking, and the object of the abduction was the speaker.

Given a Deep Structure like that, the speaker may choose from a number of possible Surface Structures, depending on the grammatical transformations her language faculty chooses to exercise. For example:

  • “I was abducted by aliens.” (Passive transformation.)
  • “Aliens abducted me.” (Active transformation.)
  • “By aliens was I abducted.” (Passive transformation, then prepositional fronting.)
  • “It was I the aliens abducted.” (Object fronting.)

And so forth. Notice in particular that the Surface Structure is not a representation of meaning; it’s just a form that meaning can take.

The great advantage of the Chomskyan grammar was that these transformations were explicitly laid out, so that given a Deep Structure, one could predict exactly which Surface Structures were permissible.

In Bandler and Grinder’s theory, Deep Structure isn’t an idea that the speaker wishes to convey; instead, it seems to be part of the patient’s subconscious model of the world. The speaker has some subconscious knowledge, such as “I am afraid of being abducted by aliens”, and only part of this knowledge is available consciously. Suppose, for example, that the only part of the knowledge that’s consciously available is “I’m afraid”. When the speaker uses transformations to create surface structures, the part about aliens is missing or generalized; so all that comes out is “I’m afraid all the time” or “Everything scares me”. The therapist can then try asking questions to elicit the aliens hidden in the Deep Structure — questions like, “Really? Everything scares you?” or “Really? All the time?” Then the patient is forced to think about what parts of the subconscious Deep Structure they’re leaving out. The patient might reveal that actually, come to think of it, she’s only afraid when aliens are around, etc.

Now, this might be a great way of doing therapy — I don’t know. But it isn’t based on Chomskyan linguistics at all. It’s something entirely unrelated to it. For example, notice that for Bandler and Grinder, the Surface Structure has its own meaning — a meaning that’s different from the Deep Structure! It’d be more accurate to say, perhaps, that Bandler and Grinder have a theory of therapy loosely inspired by Chomskyan linguistics. The relationship between the two is probably more distant than the relationship between astrology and astronomy: even though they seem to be talking about largely the same thing, their words mean different things (e.g., just as “Aries” means different things in astrology and astronomy, “Deep Structure” means different things in linguistics and NLP), they have completely different methods, and completely different goals.

Connection Between Language and Magic: Still No Neurons, but Maybe Programming and Linguistics

So the primary goals of the authors — to offer a non-theoretical toolbox of therapeutic techniques grounded in solid linguistic theory — didn’t pan out. Does that mean the books are useless? Absolutely not. Just because the map doesn’t cover the territory they think it does, doesn’t mean it’s a useless map.

The analogy between magic and NLP is made explicit in “The Structure of Magic”, where effective therapists are equated with magicians who work healing magic on the human mind. Is any of this material of any use to those who practice real magic and/or meditation?

I think so. Here are some possibilities:

  • Find your best sensory channel. According to these books (and in my own experience), everyone has one or two primary sensory channels. In these channels, we organize our experience well and communicate effectively with others. In other channels, we may not do as well. For example, my wife sounds out words in her head as she reads (despite being one of the most intelligent people I know) and has all kinds of trouble with simple visualization exercises. Her auditory input channel seems much better-developed than her visual one. Some people do better with touch, and some perhaps with other senses such as balance (my phobias and fears always seem to manifest in my stomach!). If you have trouble visualizing during magical rituals or meditation, try organizing your practice around another input channel. For example, instead of visualizing an elevator slowly going down twenty floors, imagine a calm voice slowly counting down, or imagine picking up carved wooden numbers one by one.
  • Work on activating disused channels. Consciously, but gently, try to use channels that are less developed for you. Not only does it add to your toolbox, but using multiple channels simultaneous will make for more powerful effects in both magic and meditation.
  • Examining your own soul. Every magician needs to keep close tabs on the inner life. Unacknowledged subconscious thoughts and feelings will have as great, or greater, effects on practice than acknowledged conscious ones. These books offer techniques that you can apply to yourself to being clarity to your own mental life. For example, if you are not getting the results you want in your practice, ask yourself: What are you trying to achieve when you fail to get your results? When do you fail? Where do you fail? Do you fail every time? This may seem obvious, but sometimes we don’t ask ourselves these questions because we are afraid of the answers.
  • Exploring the connection between language and magic. This is the most intriguing thing for me. The authors tried to draw a connection between language and therapy, and frankly, they failed. But they failed because they were sloppy, and because Chomskyan linguistics was extremely young, not because the enterprise itself is somehow doomed to failure. What would happen if (a) modern, more accurate theories of linguistics were used and (b) the connections were more carefully drawn?

Let me give a brief example of the last point. George Lakoff’s work on the embodied mind and metaphor is a powerful theory of language and the brain. It suggests that our language and our reason are firmly grounded in physical experience, not in abstract perfect forms or Aristotelean logic. If that’s true, what implications are there for the underpinnings of magic and meditation?

  • It means that the choice of words we use in our practice should reflect our desired effect. For example, if we want to bring money to us, we should make use of phrases such as money come, draw money, pull money rather than, say, rising income or increasing money. The latter phrases activate metaphors in which money simply increases, without respect to where or how it increases; the former activate metaphors relating directly to the movement of money towards your body, and are much more powerful.
  • It means the best path to wisdom is in learning, comparing, and creating complex metaphors — that is, stories.
  • It means almost all higher brain function is based on metaphorical mappings of a sophisticated nature. Metaphor is our best — indeed, our only — tool for understanding the world and generating change.

There is much more to be explored here, lots of stuff to uncover. Way cool.

In sum, I can recommend the books as engaging reading, and many of the techniques may work for you. But for the sake of the gods and your sanity, skip the sections on linguistics!


  1. You are in good company, another Neurolinguistic programming fan
    is Robert Anton Wilson (he writes about it in “Email to the Universe”).
    He seems like he would be your kind of author.

    Also, *love* your photo !

  2. Jeff Lilly says:

    Terry, you’re right — I think Wilson is just great. In fact, I’ve already mentioned him in this blog, in the entry about the structure of consciousness. I haven’t heard of his “Email to the Universe” — I’d better check it out!

    Glad you like the picture! It’s not a photo, though. This picture, and all the other landscapes on this site, were generated using a program called “Terragen”. It’s very easy-to-use software that makes amazingly lifelike pictures. Check it out at

  3. Hey Jeff,

    Interesting article on NLP. Having been studying it and using it actively for some time, I feel the linguistic side is actually more related to the work of Milton Erickson and Virginia Satir both of whom were modeled by Bandler and Grinder and provide a platform for much of the processes used in NLP.

    NLP uses language much like hypnosis does, to send the person deeply into their own internal processes. The reason that a person may have a different meaning on the top layer than they have on the bottom layers is that much of the real meaning for them is not in their conscious awareness.

    These processes dig out the unconscious pieces and pull them into conscious mind where they can easily be changed.

    A caveat here for those who play with the techniques, much of NLP is based on the skills of the technician. Someone skilled at NLP is trained to notice even the smallest changes in their client.

    It is an intense focus and intuition that makes a skilled practitioner so if you are just playing with a technique you may not have the same results as it is hard to walk yourself through a process, go into your internal experience and observe yourself at the same time 🙂 .

    There is power in doing this with another person who is trained, who is holding a non-judgmental space for you and who is making you, the entire point of their focus during the process. Energy is raised in that setting as well as consciousness.

    I feel that NLP works well combined with meditation and magick. I use it to help people learn to crawl into and create a communications system with those deeper parts of themselves. This is all stuff that can be enhanced upon by other practices and of course other practices can also be enhanced by NLP 🙂 .

  4. Hi Paula, thanks for giving your testimonial here. Since I wrote this article a year and a half ago, I’ve picked up a lot more about NLP here and there, and heard almost nothing but good things about it. In that respect, it’s remarkable that it began from such a flawed understanding of linguistic theory — a linguistic theory that was later abandoned by linguists! 🙂

    I can’t speak to the linguistics of Erickson or Satir; their wikipedia articles say almost nothing about their ideas about linguistics, except that Erickson appears linked to Lakoff, and then only connected with metaphor. However, the brief mentions of metaphor in the article have nothing to do with Lakoff’s ideas about metaphor, which operate at the level of sentence construction.

    I am a little worried that Bandler, Grinder, and maybe the author of the wikipedia article are appropriating famous names and terminology from linguistics in order to give NLP an aura of scientific respectability. What I’ve seen of NLP seems to have nothing to do with actual linguistics at all. If so, this would naturally only serve to undermine their respectability, at least from the point of view of this linguist. 🙂 But again, this is not to say that NLP doesn’t work — I have every reason to believe you when you say it does!

  5. Hi Jeff,

    Most people in NLP will tell you that it is a really poor name that sounded good in the 70’s 🙂

    John Grinder was a linguist…so that is probably where it got it’s start in flawed theory and all! All this to say that NLP is really about learning how people interact with their world and to that end…it is a really good tool.

    I agree that “borrowed” respectability does little to actually promote the field. It’s sad too, because NLP is a window and doorway to spiritual experiences…something that seemed to go by unnoticed by it’s founders who were concerned with proving something that is, to some degree unknowable…scientifically.

    My guides led me to NLP so I can’t say I was taking a “scientific” view of it 😉

    Much love,


  6. Oh yes, I see he was a linguist! Amazing! His use of linguistic theory was so wrong I assumed he’d had no formal training in it… How strange!

    Oh well — as you say, the whole point is academic, isn’t it! 😉

  7. “Just because the map doesn’t cover the territory they think it does, doesn’t mean it’s a useless map.”

    Or as the great horseman Ray Hunt once said, “I don’t know where I’m going, but I’m not lost.”

  8. Jeff Lilly says:

    That’s a great quote! 🙂


  1. […] to neurolinguistic programming, belief really arises out of emotion. NLP actually teaches techniques for generating beliefs out of […]

Speak Your Mind