I USED THE TERM “INCANTATIONS” in The Structure of Magic I to describe the use of language in change-work for a very good reason. Words—as occultists, philosophers, psychologists, and writers know all too well—have magical effects. When I invite clients to “sit for a spell,” the ambiguity is deliberate. I want them to begin to be open to the possibility of change—and to the fact that the change may seem magical; often, it is.
One important aspect to helping people change is making sure they feel you understand their problem, then to move them as quickly as possible from their problem state to the solution you have prepared for them. Words are the primary means by which you can help create this kind of change.
Watching Virginia Satir work, I noticed that she tended to reflect her clients’ sensory predicates—those words and phrases that signify which of the five senses is dominant at the time of speaking.
Someone might say: “I just feel everything’s getting on top of me and I can’t move forward or back. I just don’t see a way through this.” She would reply: “I feel the weight of your problems is stopping you from finding your direction, and the best route you can take isn’t clear yet…”
She did this intuitively and achieved really close connections with her clients.
On the other hand, I often observed therapists who had no concept of the sensory preferences of their clients and just spoke the same way to everybody they met. In response to “I’m weighed down by all my problems,” a less enlightened therapist might respond, “Well, you need to listen to what I’m saying so you can see some light at the end of the tunnel.” These therapists were talking a different language from their clients, and their clients felt as if they were somehow not being listened to or understood.
Couples sometimes end up in trouble by not recognizing these differences. One person—the visual partner—might express love in the form of gifts and flowers, but the other—the auditory partner—still feels neglected because the words “I love you” are never actually spoken out loud.
Once you have successfully matched the other person’s preferred sensory system, you can begin to lead them in new directions, to increase their ability to process effectively and make enduring change. We do not want the subject to stay stuck in one processing mode; this lack of flexibility landed the person in trouble in the first place.
One of my objections to the Montessori method was just this. Originally, when a kinesthetic child was identified, he was taught only by kinesthetic methods. Likewise, visual children were taught only visually, and auditory children were taught strictly by auditory methods, thereby stunting their growth and possibilities. They were stuck on one channel, whereas real learning involves crossing into other sensory channels to optimize an individual’s potential.
Expanding a client’s experience by expanding the limits of his or her subjective model is central to the methods adopted by all the truly effective therapists and teachers I have studied. Other characteristics of effective therapists and teachers include:
Among these commonalities was the kind of questions they asked. Somehow these people seemed to have the ability to ask questions that put the client on the way to recovery. When we analyzed the effective therapists and teachers, we found that they focused less on gaining more information about the possible origins of the problem, and they paid more attention to helping the client retrieve deleted, distorted, and generalized information. In this way, the client was able to reconfigure her or his internal map. The syntactic distinctions, published as the Meta Model in Volume I of The Structure of Magic, were intended to explore the under lying, full sensory representation (the deep structure) of the thoughts and utterances (the surface structure) made after information had been filtered out by the processes of deletion, distortion, and generalization. A simplified version of the model is laid out in Resource File 4 (page 311), and I suggest you spend some time studying and practicing the different patterns and their challenges. The section that follows is intended to give a feeling for what is possible with mastery of the model.
Over the years, some people have come to see the Meta Model as a form of therapy, possibly because the book included a transcript of a therapy session, identifying a client’s violations of the Meta Model together with the therapist’s challenges. But the Meta Model has nothing to do with therapy. It is a powerful, recursive, linguistic pattern used to uncover quality information. That’s why, when I use the Meta Model, I always ask for the biggest chunk of information first. I start the opposite way to that laid out in The Structure of Magic I.
The purpose of the Meta Model is to be meticulous, to ask the kind of questions that will help you find out how somebody’s problem works so that you make sure you alter just the problem, and not everything else in the person’s life.
Somebody comes in and says, “I’m depressed.”
I challenge the generalization (the Universal Quantifier) within the statement by asking, “Every moment of every single day? Even in the shower?”
They might admit, “Well, not always.”
I then ask, “So how do you know when to be depressed?”
Some people respond, “I’m depressed whenever I have spare time.”
With the Meta Model as a tool, there’s no reason to quit. I ask, “How do you know when it’s spare?”
They say, “Because my mind races…”
“Ah, the racing mind,” I go. Now I start to get quality information. I ask, “When your mind is racing, what exactly is it doing?” and this is where all the details emerge of how the subject is creating the experience: pictures going by, voices yakking away, feelings slopping from here to there, or any combination.
What actually happens with this approach is that you’re defining the experience as volitional instead of outside the person’s control. You say things like: “So if you make a picture of X, then you say that to yourself Y, then you feel Z…” This is all process, and once expressed as a process, it presupposes that the process is open to change.
If we accept the other way of saying things, “I have depression” or “The problem is my frustration,” the speaker has taken a verb and turned it into a noun (nominalization), and in so doing has also deleted information such as the fact that he’s making the pictures, saying those negative things in his head, and feeling those bad feelings.
Every sentence has a lost performative (an indication as to who is responsible for the action being complained of), and as soon as you restore that performative, you’re returning responsibility and power to the client. I use the phrase, “So, what you’re saying to me is…,” to restore the lost performative.
They might say: “I’m not happy” and claim they’ve “never really” been happy.
I can choose to challenge them by questioning the “never,” or I can say something like: “So, you’re saying to me that you can never be happy.”
They’ll say, “Well, yes.”
I’ll ask, “And how do you know that?”—because they’re making a comment about their state of mind, not about the nature of reality.
They’ll usually respond: “Well, I just know it, because…”
I’ll say: “No, no, I don’t want to know why. I want to know how you know.”
They’ll say something like, “Well, because I’ve never really been happy.”
I’ll follow up with: “Well, if you’ve never tried something, how do you know whether you like it or not? Maybe happiness isn’t all it’s cracked up to be. Maybe really happy people are actually miserable. They could be just pretending. It could all be a big con.”
Then they say, “Okay, I know because I’ve had moments when I’ve been happy.”
I say, “Ahh, so there have been moments. What was that like?”
Using the Meta Model requires a certain amount of finesse and elegance. Just asking the questions by rote is not going to get the results you want. There should always be the presupposition of change in the language you use. For example, often, as I’m bringing someone out of trance, I tell them to “go back and remember this bad feeling for the last time.” Nobody ever questions it. I say: “Have you got it?”
They say, “It’s really hard now.”
I say, “Work at it more.”
Now, whether they get the feeling back a little or a lot doesn’t matter. They’ve already accepted the presupposition that the bad feeling can and will be felt “for the last time.”
Meta Model questions are designed to gather information. You can think of the model itself as a sword that chops up meaning. It slices things out, sorting what works from what doesn’t, always moving toward whatever outcome you want.
So, whatever it is they want, your message is, “Okay, we chop away all the things that won’t get you there.”
People will tell you they want something like “being comfortable about public speaking.” The presupposition in there, right to start with, is that what they’re asking for is a good thing. You could challenge what the Meta Model calls the Universal Quantifier by asking, “Are you saying you want to fall asleep in front of your audiences?”
They’ll say, “No, of course not. No, maybe, it’s…I’d like people to admire me.”
You might respond, “For no particular reason? You want them to just to hang around obsessively admiring you?”
They’ll say, “Wow, no. I don’t want that, I want…”
You slice away the nonsense until finally they explain, “Look, okay, so, I want to be relaxed, but alert. I want to engage my audience’s attention and see that they’re enjoying themselves,” and so on.
Then they realize they’ve been going inside, seeing themselves terrified, sweating, voice cracking, everybody in the audience laughing, and you say: “Good plan. That’ll get you into the right state.”
Not only do they see that their old behavior was not a good plan, but that they’ve been doing it habitually and also unconsciously. By asking the Meta Model questions, you bring their behavior up into consciousness, make it move a little slower, then start slicing away the nonsense. It tells you everything you need to know, including what to do next.
One of my favorite cases, which I wrote about in Magic in Action, involved a woman who had psychotic episodes whenever anyone she was expecting to meet was late. She’d been in therapy for eight years, had three different therapists that I knew about, and whenever anyone asked her why she had these responses, she’d say, “I don’t know.”
But when the woman said, “I have a problem I’m too close to,” I knew the solution was to push away the pictures. She was making pictures of horrible road accidents that became progressively closer, bigger, and more detailed, until she smelled the burning metal and felt the warm blood spattering on her skin. That would scare anyone. She let me know that we needed to push the images out, make them less and less distinct until they disappeared. We did, and it worked, all in a fifteen-minute session.
I’m not trying to diagnose people with this approach; I’m trying to make the diagnosis wrong. If people come in and say they’re depressed, I want them laughing their asses off as quickly as possible, so, after that, every time they think about being depressed they burst out laughing.
I want to give them a better problem. Often I listen to clients and think: “What a sad little problem. They need something bigger and better.” They need to find the answer to questions like: How much pleasure can I stand? How much can I get done in a lifetime? How can I feel really great every time I go into a meeting or see my husband or wife?
If people don’t ask the right questions, their brains don’t learn. I always know when the questions are coming, so I throw out a better question. I say, “Stop and say to yourself, ‘It’s time to do something. What should I do?’” I just switch the Referential Index (who is saying what). It’s not elegant, but it works.
All the above examples illustrate how the Meta Model works. The questions lead us directly to where we want to go, because we’re looking at the syntax of the question, not its content. If you fall into content, you’ll drown because content is infinite. We all know how little kids going “why?” can go on forever. The fact that a psychiatrist might do that means therapy can last for years.
It doesn’t matter to me why something happened. I don’t try to read minds or encourage clients to read their own minds. I want answers that point me in the direction of making change. You have to know how to ask just the right questions, and then you have to know how to give just the right suggestions, in just the right way, so that you maximize the result that you want. Being able to move smoothly back and forth between knowing how something happens and what to do about it is what good NLP is about.
This is where the Milton patterns become so important. The Milton Model (see Resource File 5, page 316) is sometimes said to be the mirror image of the Meta Model, but while the Meta Model is applied to gain quality information, the Milton Model—derived from the patterns modeled from Erickson’s work—uses language in an “artfully vague” way to induce trance and promote change.
It’s often assumed I knew a lot about hypnosis before I heard about Milton Erickson, but when Gregory Bateson first told me about Milton, I knew nothing. So I gathered his collected works, all his journal articles, everything I could find written by him, and read it all. What I found interesting was that he was claiming to get results that nobody else said they could achieve.
I was intrigued by Milton’s claims, so I went out and got every book—literally hundreds of books—about hypnosis and read them all. I tried out everything, a lot of it on an extraordinary neighbor I had at the time. She was agoraphobic and had allergies and all sorts of things wrong, and we fixed them all. So, by the time I got to actually see Milton, I had quite a lot of experience, and I’d already analyzed his language patterns, from the journal articles and the transcripts.
It was fairly apparent that most of the people who knew about him were as mystified by him as they were by Virginia Satir and by Fritz Perls.
Virginia, who achieved consistently good results, didn’t claim to have the right approach to therapy. She just said that people could be helped more easily if all the family members were involved, rather than just the individual. Also, her ability to observe patterns and predict behavior was extraordinary. On one occasion, when I drove her to see a family whose epileptic daughter had been labeled a juvenile delinquent, she said, “Watch what happens. In the middle of this session, the girl is going to have a seizure. The moment I start talking to one or other of the family members, she’s going to fall down in a fit.” Sure enough, that’s exactly what happened. This was the sort of thing that happened around Virginia. But one of the most important qualities she had was that she was absolutely relentless. She was warm and sweet and kind, but she didn’t give up. It didn’t matter to her if it took twelve hours. She would keep working until she got the change she was after.
Virginia was an exquisite hypnotist, something she strongly denied at first. I showed her videotapes of her and Erickson, and for the first ten minutes they said exactly the same things. Virginia had nicer tonality than Milton. He sounded a little like Boris Karloff. She sounded like the sweetest person on the earth.
It was superb hypnosis, but she said it was just a centering exercise. She’d talk about people’s uniqueness, how each was the only one in the world with those fingerprints, and so on. Then I’d turn on the Milton tape, and he’d talk about the individuality of his patients, how their fingerprints were unique—the same concepts, in the same order.
It took her time to admit it, but, finally she came around, and even asked me to use hypnosis to help her with a personal problem.
Virginia had met Milton and thought he was creepy and didn’t want anything to do with him. I have to admit, I understood why she felt that way. He was in a wheelchair, having had polio twice, and was suffering from postpolio syndrome. He wore purple pajamas, induced trance, and communicated covertly more or less all the time, even when he didn’t need to. But he did it to amuse himself. Interestingly, though, despite their differences, Virginia and Milton were, in my opinion, the best at getting results.
Fritz’s work was very hypnotic, too. Telling clients to hallucinate dead relatives in empty chairs—what is that if it isn’t deep trance hypnosis?
In reality, Fritz didn’t actually have a very good track record fixing clients. Everybody was impressed with his work, but he didn’t get good results. He couldn’t get an insomniac suddenly to be able to sleep, for example, and he was very open about the fact that he couldn’t work with psychotics or schizophrenics. He only worked with “neurotics.”
On one occasion, though, he did help a client get over his impotence by having him think about his nose and then his genitals and his nose again. He couldn’t explain how it worked; he just said it was something that fit his theory. Now, of course, we know that in the motor cortex, the wiring for the muscles of the nose and the genitals are right next to each other. If you move your nose, typically your genitals will move; typically, if you flare your nostrils or move the nose up and down, you stimulate your genitals.
When the patterns I identified were first published as the Milton Model, Milton was very pleased, even though he implied they only reflected a part of his repertoire. Milton’s approach could be very complicated. He very strongly identified with the concept of “being a hypnotist” and insisted that all his clients become exceptional hypnotic subjects before they went any further.
I was more interested in how far I could push this thing called hypnosis, so I tried everything that he ever claimed you could do. This was not because I wanted to disprove it, but because if I could produce the same effects, then I knew there would be a world of things that hadn’t even been tried.
I tried things no one had ever tried before. I wanted to find out what effects could be achieved with light trance and deep trance; I wanted to see how far we could go. I have to admit that a lot of my clients went through a lot of demanding stuff so I could find easier ways of doing things.
The people who really should get credit for my work are the clients who came to me at the end of their ropes. In fact, nobody came to me first. They only came to me because everybody had given up on them. They always said, “You’re my last hope,” and I’d always respond, “Boy, you’re in big trouble then.”
But I didn’t give up. From Virginia Satir I learned to be relentless. I learned that if something doesn’t work, you just do something else. Failure is when you stop, and I never stopped.
In practice, Erickson didn’t use all the patterns that became known as the Milton Model, nor do I. Since I paid attention to Erickson, Satir, and Perls, as well as to those “ordinary” people who accomplished things by themselves, it became possible to create a technology that was universal in its application, was fast, and that anyone could learn. Quite simply, the language we use has a direct impact on the listener’s neurology. The language we use when talking to and about ourselves also affects our own neurology.
Not everybody will use Milton patterns the same way. The people who become really familiar with them will find they have certain preferences and will naturally develop their own distinctive styles.
For my part, I find temporal predicates—words that refer to time and its passage—incredibly powerful. I use temporal predicates as linkage—“when you sit here breathing in and out, then you will relax, and as you think about this for the last time…” But there are many more ways temporal language can be used.
Inducing confusion increases suggestibility—for example:
[B]efore you stop yourself from preventing the idea that you don’t know what’s coming later, it’ll be here, but before we start to continue with what isn’t important about what you don’t know, you’ll find that you’ve just begun to go backwards, because the past is just a future moving by now…
This passage demonstrates how language patterns can be layered. Aside from the temporal predicates, that last sentence is stacked with ambiguities—words and phrases that could have more than one meaning, leaving the unconscious room to explore alternatives that have not been explicitly stated.
Another reason I regard temporal predicates as particularly important is to make clear the very important distinction between the past and the future. The best thing about the past is that it’s over. When people don’t deal with the past as if it’s over, then they’re not free to go into the future. That’s why I particularly love the ambiguity that “the past is just a future moving by now…” (I suggest that you reread that sentence very carefully to find out for yourself how many meanings it contains.)
I often talk about people being angry or sad or depressed “for the last time.” I like what are known as “semantically dense” predicates, something linguistics spends a lot of time discussing. For instance, one doesn’t lurk up to somebody openly. The verb “lurk” has all kinds of connotations that don’t need to be stated, so when you say that somebody is walking around the edge of a crowd, as opposed to lurking around the edge of a crowd, the semantically denser phrase has greater impact.
Temporal predicates—words like “last,” “first,” “after,” “again”—all have semantic density. Phrases including the word “when” (“when you start to do X, you’ll find something important”) and “next” (“the next time you see him, you’ll feel Y”) really allow you to aim posthypnotic suggestions to maximum effect.
I think of temporal predicates as targeting devices that allow you to place feelings, amplify them or diminish them, with great power and precision.
Temporal predicates, of course, are directly connected to presuppositions. Presuppositions literally “presuppose” or assume that something is present, even though they are not explicitly stated. A question such as, “When you get up, could you close the door?” contains a number of presuppositions: that the listener will get up, that there is a door, that he is capable of closing the door, and so on.
Many syntactic environments for presuppositions are based on temporal predicates. The “when” in the previous example is a temporal predicate that supports the presupposition. I find these to be extremely powerful, especially when you talk about doing something “for the last time,” or about feeling something “never again and again and again.”
There are also wonderful, simple, and effective words like “stop.” Most people don’t think of “stop” as a temporal predicate, but when I see people beginning to go into a behavioral loop that’s going to run ad infinitum, where they start to get a bad feeling or a panic attack, I say to them, “Stop”—and, amazingly, they usually do.
Add to that a phrase such as “back up,” and you have even more effective tools. When someone is sitting down, there’s no way to physically back up, so when you say, “Stop. Back up and feel something else this time,” they know at a deep level what to do.
Another word that is temporal in nature is “new.” “New” implies that you’re going to do something in the future so “this old feeling that’s going past isn’t going to be as satisfying as when you find new feelings coming…now.”
“Now” is one of the most powerful temporal predicates in the hypnotist’s repertoire. People, especially in altered states, can be very passive, so you have to tell them what to do, when to do it, when to start…and now, of course, is a good time. If I tell people to “go deeper,” it doesn’t mean they will. I tell them exactly when to do anything I want them to do: “Your arm will drop…now”; “In exactly two minutes you’ll find these thoughts coming into your head, now, and then you’ll find…”
Ambiguity is a useful pattern when working with somebody who has a suspicious conscious mind and doesn’t trust himself. Then I’ll talk “through” them to their other parts, trying to come in from the back door to the front door, instead of the front door to the back. Of course, if I have the subject’s cooperation, I’ll use it. I’ll get the conscious mind and the unconscious mind doing the same thing. The more you can line up a person’s resources, the better off you are.
The categories known as punctuation and scope ambiguities need special attention. Not only are they effective in themselves, but they are also modified by temporal predicates. “Time and again and again you’ll start to have old feelings disappear”; “Those same old feelings will come up for the last time just before you feel them now disappearing…”
These patterns are very hard for the conscious mind to follow, but very easy for the language-processing centers of the brain to compute. I don’t know how many times I’ve given people suggestions, and they looked at me and said, “What?”…and then carried them out to the letter, at precisely the right time, because they were given specific temporal markers.
Now, take a minute or two to find a new idea…
Milton used the phrase “Your unconscious now” (“you’re unconscious now”) many, many times. It’s a great ambiguity, but as soon as you slam that temporal predicate after the word “unconscious,” it also becomes a command. “Your unconscious now…wants new ideas,” “Your unconscious now wants to know even more unconscious now…You’ll see that you’re not doing what you can see the future coming now…”
All of those kinds of temporal phrases give you great room to put content on either side. It’s about deciding a direction and aiming where you want things to go. What you’re doing in hypnosis is leading someone’s consciousness down a certain path, and you have to decide whether that path leads into their past or their future. Some things you want behind them and some you want in front. Some you want gone forever.
Forewarned is forewarned…and the more warned you are about where you’re not going…you need to have signs in your mind that say, Stop, go back, you’re going the wrong way. In the United States, they put those on freeway on-ramps so you don’t go on the wrong one and end up going against traffic. I install them in people’s minds. I say: You need a sign in your head that says, Go back, you’re going the wrong way!
Now, stop, go back, and remember that idea you just thought about, only just get to the sign at the entrance. Bad idea. Go back. You’re going the wrong way…now. And then see the signs of where you should go. Pleasure ahead. Happiness coming. Choices ahead. Past behind. Leave it behind, now, so when you go ahead of time—because it’s not enough to be in the now—you need to be ahead of the now, because the future is coming, the past is behind, so never, yeah, never do never again. Never forget what you shouldn’t remember. And always remember what you shouldn’t forget…now. And then you’ll do it correctly. Because, once again (I love that “once again”), you’ll find tomorrow is much better.
Yes to day (I love that one, too. That’s full of logical ambiguity, “yes to day”). And when it comes to hope, yes to day has no bearing. Now…
Notice how densely the language patterns are stacked. When you have temporal predicates and presuppositions, and when you stack presuppositions—at least three at a time—it becomes extremely difficult for the listener to track consciously, so it produces a very strong effect on the listener’s unconscious.
Another pattern I’m particularly fond of is “the more, the more” pattern. I use that one all the time, especially with negations stacked one on top of the other. “The more you try to stop yourself from preventing what you know that you don’t understand, the more you will, because, as you try to continue to not do something you won’t be able to not see what’s going on.”
The purpose is to overload the unconscious, and once that happens, the doors open up and you can flood in the suggestions.
I often say that I’m not a hypnotist so much as a “hypno-ranter.” Where most people are providing gentle, nondirective suggestions, I’m slamming things in from every side, and every way that I can.
Speaking to the unconscious processes inside somebody with semantic density is an art form. It’s almost like being able to write good poetry, but it doesn’t come from nowhere. It’s not an innate talent. It’s something you develop, and the way you develop it is through practice.
I recommend that you spend two days on one kind of syntactic environment and the next two days on another. You can refer to Resource Files 4 and 5 (pages 311 and 316) for further explanation and inspiration, but to be able to generate language patterns without needing to think about them, you should write down pages and pages of each pattern. Reconfigure your brain so that it all becomes familiar and easy.
If you don’t have a lot of examples of what makes things different, it’s very hard to make yourself familiar with it. Hypnotic language patterns, hypnotic states—these are the building blocks. If you didn’t know all the letters of the alphabet it would be very hard for you to write anything.
People often consider me to be a very complicated person. It’s true that I know a lot of really complicated things, but when I work with human beings, there’s nothing complicated about it at all. I have broken things down for years and learned how they work, and then I’ve practiced putting them into effect. I studied language patterns so that I can automatically and unconsciously generate them in many sophisticated forms. I don’t need to think about them anymore. I just do it, while keeping my eye on where I want to be.
These are the things that set people free.
Exercise 1: The Meta Model
- Refer to Resource File 4. Begin to practice noticing Meta Model patterns, spending two days on each. Pay special attention to the language you hear, noting the violations that occur. Television interviews with politicians are a rich source of Meta Model violations.
- As you become more familiar with each pattern, jot down some of the challenges you would use in a real-time situation.
- Working with a partner, discuss a real or imaginary problem. The listener notes Meta Model violations and challenges them, always seeking to recover information that has been deleted, distorted, or generalized.
- Change places and repeat.
Exercise 1: The Milton Model
- Review the examples given in Resource File 5, then create at least twenty of your own.
Exercise 2: The Milton Model
- Decide on an outcome you would like for a client. Choose three to five Milton Model patterns, and create a conversational induction by linking the patterns with conjunctions or temporal connections. Repeat the pattern three times, so that each induction comprises between nine and fifteen examples of hypnotic language.