Reframing. Neuro–Linguistic Programming™ and the Transformation of Meaning

Creating a New Part

One of the questions that we have asked over and over again since the beginning of our dealings with the field of psychology is «What is it about an experience that makes it therapeutic or not therapeutic?» Every school of therapy has within it certain elements which lead to change when used by some people, and don't lead to change when used by others. When used by a third group, those elements lead to change which is not profoundly useful. As far as I can tell, the ways you change people into behaviors which are not useful are not really different from the ways you go about changing them into behaviors which are useful. The kinds of techniques that are used by well–intentioned parents, probation officers, and teachers, to lead people into behaviors which will actually cripple them for the rest of their lives, are powerful and effective mechanisms of change.

This morning we want to teach you a third model of reframing: how to create a new part. Parents, educators and well–meaning psychotherapists don't create new parts as explicitly as I'm going to teach you to do. They mix the pieces up, and they do it over a longer period of time. However, those of you who are therapists will recognize the elements readily. This model has more steps to it than the six–step reframing model, and it's designed to accomplish something entirely different.

The presupposition of the six–step reframing model is that somebody has a part that deliberately stops her from doing a behavior, or a part that makes her do a behavior.

Yesterday afternoon we dealt with a second logical possibility: that there are two or more parts, and each of them is doing exactly what it is supposed to be doing. Their intentions are positive and their behaviors are appropriate, but when those behaviors overlap, they produce an unwanted condition such as insomnia. You have a part that takes care of business and methodically plans out everything, and you have a part that wants to go to sleep. When one part of you starts to go to sleep, then the other part goes «Oops! You forgot about X! What's going to happen if you don't do this?» The other part says «Don't worry about it now. Let's sleep.» However, you didn't find a solution, so as you begin to drop off to sleep, the other part says «But if you don't, Y will happen.» The negotiation model is adequate to deal with situations like that. You negotiate between the parts so that they work more cooperatively.

This morning we want to explore a third logical possibility: somebody doesn't do something simply because there isn't any part of her that's organized to do that behavior. There is no part actively stopping a behavior, and there aren't two parts interfering with each other. She has lots of other parts that work. Consciously she desires a particular outcome; however unconsciously she really doesn't have a part that can carry out that particular behavior.

All the other reframing models change a response, and that new response triggers a different sequence of behavior. For example, in verbal content reframing you just change the response and assume that it will fire off more useful behaviors. Of course you need to check to be sure that assumption is correct.

In six–step reframing, you change the response, and you ask the client's creative part to go on an internal search to find specific alternative behaviors. You anchor those behaviors into the appropriate context by future–pacing, and do an ecological check. When you negotiate between parts you assume both parts have appropriate behaviors already, and you just need to provide a way for them to sequence when they do their behaviors, so that they don't interfere with each other.

Content reframing, the negotiation model, and six–step reframing all presuppose that either 1) alternative behaviors already exist, or 2) some part can easily organize itself to carry out behaviors that will be appropriate. Those are very useful presuppositions, but they aren't always true. If I put one of you alone in the cockpit of a Concorde SST, you could be perfectly calm and alert with no parts interfering with your behavior, and still not know how to fly the plane. You just don't have the appropriate behaviors organized to do that. You need to go through some kind of learning process to organize and sequence those skills. That is the kind of situation in which you have to create a new part to do a specific behavior, and that is what most education and training is supposed to do.

A few years ago we were doing a workshop up in the Northwest, and one woman in the seminar had a phobia of driving on freeways. Rather than treating it as a phobia, which would have been much more elegant, we did a standard six–step reframing. We don't recommend that you use reframing with phobias, because usually your clients will get the phobic response as a signal. Once they've collapsed into the phobic response, it's very difficult to do anything else with them. However, we were demonstrating reframing at the time, and decided to demonstrate that it's possible to do reframing with phobias.

We said to this woman «Look, you have a part that's scaring the pants off you when you go near freeways. Go inside and reassure this part that we know it's doing something of importance, and then ask if this part is willing to communicate with you.» The woman got a very strong positive response, so we said «Now, go inside and ask the part if it would be willing to let you know what it's trying to do for you by scaring the pants off you when you go near freeways.» The woman went inside, and she reported «Well, the part said 'No, I'm not willing

to tell you.'"

Rather than go to unconscious reframing, we did something which may sound curious but it's something I do from time to time when I have suspicions, or what other people call intuitions. We had her go inside and ask if the part knew what it was doing for her. When she came back outside, she said «Well, I … I don't… I don't believe what it said.» We said «Oh, yeah? Well, go ask if it's telling the truth.» She went inside and then said again «I don't want to believe what it said.» We asked «Well, what did it say?» She said «It said it forgot!»

Now, as amusing as that sounds, I've always thought that was a great response. In some ways it makes sense. You are alive for a long time. If a part organizes its behavior to do something and you really resist it and fight against it, the part can get so caught up in the fight that it forgets why it organized its behavior that way in the first place. That's a real possibility. I don't know how many of you have ever gotten in an argument, and in the middle of it forgot what you intended to do in the first place. Misers are like that. They've forgotten that money is only useful if you spend it now and then. Parts, like people, don't always remember about outcomes.

Rather than going through a lot of rigamarole at that point, we said Look, this is a very powerful part of you. Did you ever think of how powerful this part is? Every single time you go near a freeway, this part is capable of scaring the pants off you. That's pretty amazing, you know. How would you like to have a part like that on your side? The woman said «Wow! I don't have any parts like that on my side!» So we said «Go inside and ask that part if it would like to do something that it could be appreciated for, that would be worthwhile, and that would be worthy of its talents.» Of course the part went «Oh, yeah!» So we said «Now go inside and ask that part if it would be willing to be responsible for being sure that you are comfortable, alert, cautious, breathing regularly and smoothly, and in sensory experience when you go on a freeway entrance ramp.» The part went «Yeah, yeah. I'll do that.» We then had her fantasize a couple of freeway situations. Previously she had been incapable of doing that; she would go into a terror state, because even the fantasy of being near a freeway was too much for her. When she imagined it this time, she did it adequately. We put her in a car, sent her out to the freeway, and she did fine. She drove happily for three hours and ran out of gas on the freeway.

Now this made me curious. I thought «If you can have a part hanging around that's not doing much, and you can give it some other job, you can probably build a part from scratch!» When I thought about it, I realized that's what Transactional Analysis does. TA goes through a rather laborious procedure to build three parts—parent, adult, and child. The Michigan TA people build nine parts. If you can build nine, you can probably build any number. If you can build a «critical parent» to torture you all the time, you ought to be able to build just about anything.

When you start thinking about it, most therapies teach you how to have your parts organized. Gestalt builds a topdog and an underdog. Psychosynthesis is a little bit more creative about it: They've got a big circle, and you get to have a whole bunch of parts inside. However, they all have to be famous people; there are no unknown parts.

Most of the time when parts are described, they are described not in terms of what they do—their function—but in terms of how they do it—their behavior. If you have studied the psychosynthesis model or the TA model, you know that people usually describe, isolate, and create parts in terms of how the parts behave. So for example, if you go through a Satir parts party, you might have a «stupid» part—a part that makes you act stupid. At the end of the party, rather than being a «stupid» part, it would become your «ability to learn at your own rate» or your «ability to ask questions» or some other positive behavior. The behavior goes from being something negative to being something positive. However, it is still a behavior that is not clearly tied to an outcome. This is a very important difference. We build parts to achieve outcomes. The parts that are created through the random processes that people use in therapy usually achieve behaviors rather than outcomes.

Every therapy I've ever studied has within it some way of building parts. Some people don't have an unconscious mind until they go into hypnosis. If you believe that the «unconscious mind» exists a priori, then one day you're going to hypnotize somebody and when her conscious mind is gone, you're going to be all alone! That has happened to me. You can't assume that everything is there. Sometimes a person has all her marbles in her conscious mind. Sometimes a person doesn't have much going on in her conscious mind, but has a very well–developed unconscious entity that is a single organized unit. Sometimes that has happened through therapy and sometimes through experience.

No matter how parts are created, people have a tendency to describe how a part behaves, rather than to describe the behavior in relationship to outcomes—what that behavior does for them. At one of my first workshops for TA people, I said I believed that every part of every person is a valuable resource. One woman said «That's the stupidest thing I ever heard!»

«Well, I didn't say it was true. I said if you believe that as a therapist, you'll get a lot farther.»

«Well, that's totally ridiculous.»

«What leads you to believe that that's ridiculous?»

«I've got parts that are totally useless. All they do is get in my way.»

«Well, name one that's useless.»

«No matter what I decide to do, I have a part that tells me that I can't ever do it, and that I'm going to fail. It makes everything twice as hard as it needs to be.»

«I'd like to speak to that part directly.» That always gets a TA person, by the way. Talking directly to a part isn't in the TA model. Then if you look over her left shoulder while you talk to that part, it really drives her nuts. It's also a very effective anchoring mechanism. From that time on, every time you look over her left shoulder, that part knows you're speaking to it.

So I said «I know that that part of you does something very important and is very sneaky about how it does it. And even if you don't appreciate it, I do. Now, I'd like to tell that part that if it were willing to inform your conscious mind about what it's doing for you, then perhaps it could get some of the appreciation that it deserves.» Then I had her go inside and ask that part what it does for her that is positive. It came right out and said «I was motivating you.» When she told me that, she added «I think that's weird.» So I said «Well, you know, I don't think it would be possible for you, right now at this moment, to come up here and work in front of this entire group.» She immediately stood up defiantly, walked up to the front of the room and sat down.

Those of you who have studied strategies know that this was a demonstration of the phenomenon that we call a «polarity response.» This part of her was simply a Neuro–Linguistic Programmer who understood utilization. It knew that if it said «Aw, you can go to college; you can do it," she'd respond «No, I can't do it.» However, if it said to her «You're not going to be able to cut the grade," then she would say «Oh, yeah?!» and she would go out and do it.

I began to discover that no matter how you organize yourself, or what parts you build, if the model that you use to think of parts is tied to how they behave, then 1) You don't do them justice, and 2) You might be right, which would be dangerous. If you really had a part that didn't have a positive function—it was just critical or destructive— then what can you do? Exorcism?

There is a guy in Santa Cruz who exorcises parts. The exorcism is terrible; it takes a long, long time, and has some unfortunate consequences. This man has «discovered» an epidemic of multiple personalities in this country that no one else has noticed! He doesn't even begin to suspect that he is creating them.

I wouldn't recommend exorcism as an approach. I would rather tie parts to outcomes, whether or not they were tied together originally. If you act as if they are, they will be. Once you have an outcome, you no longer need to exorcise a part. You simply give it new behaviors.

If someone doesn't have a part to do something, you can create one, but you need to be sure that the part is designed to achieve a specific outcome. If you are not able to open doors, you can create a part that opens doors. It sounds simple; it's actually somewhat complicated. However, it's something that you do all the time. All of you have parts which you managed to make somehow or other. All the things we do explicitly with parts and reframing are things that people do anyway. These are all naturally occurring processes.

I think there's a tendency for human beings to organize themselves in terms of outcomes that are contextual. A man behaves differently with his wife than with his colleagues at work; he has an entirely different set of analogue behaviors in order to get different outcomes. That used to be called «role theory," and I think role theory was on the right track in some ways. However, therapists got stuck trying to prove that that's all there was.

Many of B. F. Skinner's students have gotten stuck in the same way. They said that since Skinner didn't look in the «black box," there wasn't anything in there anyway. Skinner didn't say «There's nothing in the black box»; he said «I'm not going to open it.» Those are two very different statements. Skinner's students took the connotations of his statement to mean there was nothing in there anyway. That is not the case, and I do not think, from reading his writing, that Skinner intended that. However, we all know how some people are: if they don't see something, it doesn't exist.

In order to build a part to achieve a specific outcome, the first consideration is to identify a «need.»

Woman: Could you distinguish need from outcome? I don't understand what you mean by need in this context.

Well, that's why I put it in quotes. What you're going to do is find an outcome. What your client is going to tell you is that she has a «need.»

The tricky part about this is to build a part that won't interfere with the rest of the person's outcomes. If there really is a part that stops her from doing something, and you build a part to do it, guess what's going to happen? WAR. To prevent this, we have built into the model that all the parts of the person that don't want you to build the new part become allies during the design process.

The first thing you do is identify whatever «need» it is that you are going to build the part for. For example, a woman might come in and say «Well, you know, I've been on lots and lots of diets and I never seem to lose weight. I'm just much too heavy, so I want you to put me in a trance and make food taste bad.» If she really wants that, I would recommend that you send her to one of those Schick clinics, where they will put big cakes in front of her and shock her. If she smokes, they will put her in a room full of cigarette butts and make her drink ashes, and all kinds of wonderful things.

That's a way of building a part that stops you from doing certain things. However, it doesn't take into account the secondary gain—the outcome of the problem behavior. That makes it a very difficult way to stop behaviors. It is an experiential way of going about it, and it will Work insofar as it's reinforced. Sometimes after a period of time, when the part that you have developed discovers that you're not going to get shocked any more, then it won't care if you smoke. So you might have to go back at a later time and repeat the procedure or do something else. That's a problem with building parts in that particular way. However, don't underestimate that approach, because it works. It seems a little severe and it doesn't work with everybody, but it does work; that's an important consideration. It's important to understand what goes on when people change, and to make up a metaphor or a lie to describe it that enables us to be able to make changes more elegantly.

Let's go back to our overweight client. Her expressed need is to «lose weight.» However, if you build a part whose job it is to lose weight, what's going to happen when she loses weight? She will lose some more! She may become an anorexic! So if you opened up a weight clinic and built parts to lose weight, you would end up needing another clinic down the street for anorexics. There you could build eating parts, and you could have the client switch back and forth every six months. There's nothing in her stated outcome that has anything to do with stabilizing weight.

Most people really don't understand substituting symptoms. There's one school of thought that says «Well, if you use hypnosis, then you will get symptom substitution.» My response is «Bravo! Let's deliberately substitute something and have it be something useful.»

Years ago a man wrote an article in which he described making cigarettes taste like the worst thing he could think of—cod liver oil. The client he did this with quit smoking, but he became a cod liver oil junkie! He carried a bottle of cod liver oil in his coat all day. I guess that's better than smoking. I don't know the ramifications of overdosing on fish oil. It sounds disgusting to me. I prefer to substitute symptoms that are positive.

So the really important question is «What is it that you are going to do in terms of an outcome?» If somebody comes in and says «I want to quit smoking» and you make the outcome no cigarettes, then the way you organize that person's resources to suppress that activity can have lots and lots of other outcomes that are not positive.

The question is «How can you conceptualize change work so that you avoid undesirable side effects?» When somebody comes in with a weight problem, what part are you going to build? In other words, what is going to be the outcome of the part that you build? At the moment, her need is to lose weight. But how can you do that and not have her end up an anorexic?

Ann: You could set a specified weight that she wants to weigh, and not let that part function when she gets under that weight.

Well, yes. We can put semantic conditions on when the part is to be active and when it's not. You could have the part begin to respond every time she weighs more than a certain amount. However, parts don't like to be inactive.

Man: You could get all parts to agree on the same outcome.

Try it some time! I'm serious. If somebody comes in and wants to lose weight, you try to get the part that likes candy to agree to that. His parts may all say «Well, that's a groovy outcome.» But if you get all his parts to agree that it's a great outcome, it still won't take him there. What the parts object to is the process of getting there.

Man: Could they generate alternatives?

You can have them do that, but then you're using a different reframing model. Then you're saying that the problem is a result of the interaction of the parts you have now. You could use the six–step model to do that. However, it's not very elegant, because then you have to go in and deal with a huge number of parts. The question is simply one of expedience: if you were only going to build one part, what would it do? I want you to make a distinction between the outcome—what you want to be sure happens—and the behaviors or procedures that the part uses to get the outcome. They are both important, but now I want you to specify outcomes.

Man: You need to make it more versatile, so that it can do more than one thing.

OK, but what is it going to do? What is its job?

I don't want to get you off the track. If we build a part whose function is to have somebody weigh a hundred and five pounds, that will work. That's great. That's a well–formed outcome. Now I want to ask «What are other outcomes that will work equally well?» There are lots of right answers to this question. The important thing is that you learn how to conceptualize them. Ann was on the right track. She said «These are what we don't want; this is what we do want. This is one way of getting only what we want.» The key question you have to ask yourself is «Will this give us ONLY what we want?»

Man: You could put the part in charge of «health» or «attractiveness» or some superordinate structure that includes weight.

Woman: How about a «central eating control» that takes all those factors into account in the process of deciding how much she should weigh?

Woman: I think you have to take all those needs which that part intends to gratify, and satisfy those needs in different ways.

Well, that's all true. The question I'm asking is «What is the part going to do?» If we have a part whose job is to be in charge of overall health, and we include maintaining a certain weight in that, then will we do only that? The magic word is «only.» Sometimes it may be advisable to do a lot more than the client requested. But right now let's talk about limited therapeutic change. Your answer is an accurate answer, and may be a better answer in experience, when you are treating clients. However, each job that apart has makes it harder to install that part. I want you to keep that in mind. Every extra outcome that a part has makes it more complicated for that part to function. The more limited its outcome, the easier it is to install apart. Sometimes it is better to make the extra effort to install a more complicated part in order to get a better result in the end. A part that keeps someone at a certain weight is going to be a lot easier to install than the kind of central coordinator that you're talking about, because the coordinator will have to have a lot of knowledge about what it means to be healthy, etc. It will also have a lot more behaviors, so you're more apt to get objections from other parts.

Man: How about installing a semantic cue for eating, and a motivation strategy to get her to eat only then?

Well, you will always be doing that. Ann suggested that the semantic cue be a particular weight. However, I want you to talk about it differently today. Part of the game we play is how you can change how you talk about experience so that you can make changes in different ways. You're still playing the strategy game. There's no such thing as a strategy, and there's no such thing as a part. The question is «How can we talk about it differently and be able to do different things?» If you forget that, then I recommend that you build a part that reminds you. «The map is not the territory» and that's not true, either.

Man: How about changing the extra weight to happiness?

I beg your pardon? A part whose job it is to dissolve weight and turn it into happiness?

Man: As long as we're dreaming.

Sure. OK, but is that going to do only what the person wants? You see, there's a big danger in what you are suggesting: six months later this thirty–pound person is going to walk in the door smiling and saying «You're the best therapist I ever had!» Either that or she's going to come in with both arms gone, and she's still going to be fat. She'll say «I feel great … but I've got this small problem… .»

I want you to listen very carefully to your definition of a new part, because in experience, that's the kind of thing that will actually happen. I think that anorexics are made by well–intentioned people, although not necessarily by therapists. Parents often give a young woman lots and lots of messages that make food such a negative anchor that she throws up when she tries to eat. Through the positive intentions of parents, the daughter ends up becoming anorexic.

I'm recommending that as a therapist you be very cautious about specifying outcomes. The more carefully you specify exactly what a part is going to do, the less you'll get objections from other parts about having it exist, and the better it will actually be able to function. If a new part is poorly designed, the other parts will be more likely to wipe it out. If we build a part that is going to take weight and turn it into happiness, that part's going to get annihilated! All the other parts are going to do an exorcism. What about the part that likes to eat candy? It's going to pull out its samurai sword, sst—whacko, and that's it. I'm asking you to conceptualize definitions for the purpose of making installation easy, effective, and useful.

Man: How about a traffic control?

You'd better be more specific than that. This is not a metaphor seminar.

Man: A part that would sequentially direct other parts to do their thing to get the desired outcome. Most overweight people know how to gain weight and lose weight. Maintaining it is the difficulty.

So you would have a part that's in charge of maintenance, for example?

Man: It would provide directions for other parts, and say «Now you do this, and you do this, and then you do this, and now it's your turn.»

OK. That's certainly a possibility. What other outcome could you specify for a part that you could build to take care of this problem?

Bill: I'm thinking about a client of mine who eats mainly in the evening when she's alone and bored. I want to build in a part so that whenever she is alone and begins to feel bored she will immediately generate several interesting activities she can engage in, so that she'll do those instead of eating.

OK. That's a strong possibility, assuming that your information is accurate, namely, that eating at that time produces the unwanted weight. Then the first question you have to ask is «Does she already have a part whose job it is to entertain her when she gets bored?» She might have one already, and the way it does that is by stuffing candy down her mouth. Then all that part needs is three other ways to entertain her. The six–step model would be adequate to do that. That is one possibility. Or it may be that she doesn't have a part to entertain herself, and it would be appropriate to build one.

Bill: She has had a lobotomy, which raises some interesting problems.

It could. I don't think parts get cut out that way, though. I think they become subdued.

Bill: But it does raise some anchors in her mind about what she is incapable of doing or thinking.

Well, all you have to do is produce research which proves that it's possible to make the changes you want to make. I'm sure you can come up with lots of studies and dates.

Bill: I've just remembered a whole volume.

Yeah. It's known as «instant research.» For some clients it's very, very valuable.

Let me give you another problem. Let's say you are using the six–step reframing model. You ask «Do you have a part of yourself you consider your creative part?» And the person says «I don't know.» And you say «Well, go inside and ask if there is a part of you that can do things creatively.» She goes inside, and then she comes out and says «Nothing happened.» And from your outside observation, nothing happened. There are two choices at that point about how you create a creative part. One is to act as if you received messages from one that was there. If you congruently convince them of this, they'll build one on their own. The other choice is to officially build them a part because they don't yet have one that can perform that function.

What other kinds of contexts would be appropriate for building parts, rather than reframing ones that are already there? Give me some examples of when building a part could be more useful than messing around with the parts that already exist.

Man: Some kind of a history that the person has never had, never experienced in his life?

That's not an example. Examples are content–specific, so that when you name them, people can go «Oh, yeah.» What you gave me was a class, and that's a different game. Give me an example of the class you were talking about.

Man: Let's take someone who has never had a satisfying sexual experience.

OK. What are you going to build a part to do? Do you want a part to make them feel OK about that, so that every time that they're sexually dissatisfied, they say to themselves «I'm OK, I'm OK»?

Woman: Have them imagine one, build it in the imagination of the person.

That is how we are going to go about building parts. You're back to procedure again, instead of outcome. I'm not going to teach you how to build parts until you know what you are building. It's an old rule of mine. Give me some examples of what you would build parts for.

Woman: If someone is born with a coordination problem, you could give them a strategy for being coordinated by copying somebody else.

OK. I'll accept everything except the preface about being «born that way.» There are many people who could use a hand–eye coordination part who were not «born that way.» They just never developed a part that had anything to do with being coordinated. Even if there's a physical impairment, it still might be appropriate to build such a part.

Bill: Let's say I have an army brat who was raised overseas and who has not had the same experiences that most of his peers have had. I want to help the kid acquire certain kinds of social skills. I could build a part to teach him how to listen carefully to people around him and to build a new history from what he hears people talking about.

What's the outcome of the part's behavior?

Bill: The outcome of the part's behavior is to teach him to speak congruently about things that he has never experienced, so that he can build social bridges and not seem different.

That's still behavior. What's the part's function?

Bill: To help him increase his social interaction with other people.

OK. That's the outcome. I can think of situations where that would be very appropriate.

Man: You could build a part that would motivate you to explore new things, risky things.

A part whose job it is to actually instigate behavior in contexts that are risky. We call that a «chutzpah part.» I can think of many people who need one of those.

Man: It seems like you're building parts into us all the time. For instance, you build parts in us to observe sensory experience and translate it, and to understand reframing.

Of course. Sure.

Woman: I'd like a part to hear pitch. The outcome would be to be able to sing on pitch.

Hear, hear. Give me some more examples of where you would use this model as opposed to the other reframing model.

Man: Build a part to learn how to do any physical activity, like roller skating or ice skating or any other sport.

Sure. It sounds like strategies, doesn't it? All right, let's go back to therapy now; we're drifting off into the land of generativity. I think it's nice that your tendency is to become generative; however, this seminar is about therapy. The question is «What problems do people bring you as therapists that this model is going to be appropriate for?» If you can't think of anything, there's no reason to teach it to you.

Woman: You could use it working with a child of a missionary who's never had a stable home environment and doesn't feel like he belongs anywhere.

What would that part do?

Woman: It would allow him to feel at home wherever he is.

Sure, OK.

Woman: How about building a part to stand up and be assertive?

Sure. What do you think assertiveness training is all about? «Now we're all going to build the same part here together.»

Man: A part to know when to get out of unproductive relationships.

Hear, hear!

Woman: A part to tell people what need their present operation enhances. And they would know why—

Wait a minute. «Operation» is an anchor for me for something other than what I think you are talking about. That was a «cutting» remark, so please rephrase it.

Woman: A part to tell someone why the activity he is involved in is satisfying to him.

A psychotherapist part? «You are happy now because …»

Woman: Then they would know why they are eating, what need the eating is satisfying, and then they could substitute a preferential activity.

Let me question you on this a little bit, because this is something that therapy has been trying to do for years, and I don't agree that it's a useful outcome. What's the outcome of installing a part that tells people about their ongoing behavior while they are doing it? There's one outcome of that that I think is absolutely disastrous: there are always two of you there at all times—one of you doing something, and one of you talking about it. That is called dissociation.

Above and beyond that, there's an even worse outcome of that kind of awareness, which is that you don't have very much external sensory experience. The outcome of having a part that constantly monitors your behavior is that you will always be on the inside, monitoring your behavior. You won't know how the world is responding to you. You will be there talking to yourself about why you are having this conversation and why you feel bad. But you'll never see the external behavior; you'll be too caught up internally. That kind of part has important limitations which should be considered.

You can build a part to do that, and in fact, many therapists already have one. They come to my groups and I say «Now, I'm going to put this person up here, and I'm going to touch him on the shoulder and his skin color is going to change.» The people with these parts go inside and say «Well, I'm feeling threatened by this. Why am I threatened right now?» Then I ask «What color did he change to?» They come back out and say «I didn't see anything.» The problem is you can't see or hear much externally if your attention is inside monitoring your behavior— whether you do it visually, auditorily, or kinesthetically.

When you're caught in a loop, you might want to have what we call a meta–part that temporarily dissociates and takes an observer position and says «Hey, what's going on here right now?» That part's function would be to get you out of loops. But the only time it would engage itself is when you are in a mess; it would not analyze at all times. If you put that constraint on it, then you begin to get a more useful outcome. The importance of thinking very carefully about outcomes is that you can succeed very well at installing parts that will completely drive people bananas in unuseful ways. So when you consider installing something, I want you to ask yourself «Well, what's the logical outcome of building this part? Is this really what I want to do, or is there something else that I have in mind? How can I be more specific about my description, so that when I build it, I get something that approximates what I want?'

Let me generalize the idea of a meta–part. A meta–part is only operational at certain times, and the contextual cue that triggers its functioning is usually based on how other parts are functioning. For example, it could be a part that comes into play only when you feel stuck, dissatisfied, or doubtful. Its functioning could also be cued to an external stimulus like a time of day, but if you do that, it may interfere with whatever else you happen to be doing at that time. So it's usually better to have it triggered off by an internal state—a feeling of being in conflict, indecision, or something like that. You can specify that whenever two parts get into a conflict, then the meta–part goes into action.

A meta–part is kind of like an amnesia state waiting in the wings to be fired off. Within the meta–part is a program, a formal set of procedures, that comes out linearly. It's like a computer sub–routine more than anything else. «If parts disagree, then do X.» The meta–part operates and modifies the disagreeing parts. It operates on the other parts, but is only functional in response to a cue. The procedure that it uses is usually formal: it could do six–step reframing, it could do content reframing, or it could just give you amnesia. There are lots and lots of possibilities for what a meta–part can do. It's a part that influences other parts to keep them from being in conflict with each other, or keep them from doing something that makes you get arrested, or whatever.

One way of thinking about a meta–part is that it is a mechanism to build a response. Another way to think about a well–functioning meta–part is that if you go into a calibrated internal loop that is not useful, that state becomes an anchor for a procedure that elicits a response that will get you out of the loop. That is closer to how I think about it than as a part. The notion of parts is a good pace for most people's experience, but for me there is a bit too much anthropomorphism in the notion of parts. You can think of a meta–part as a part that makes a distinction and then kicks into a procedure that can take you somewhere else.

With a couple, you can build a part in one of them that operates only when they argue. This part recognizes that the reason they argue is because they want things to be better. Rather than going in and negotiating with all the parts that feel right about things and argue, you can build a part that recognizes that they are now making themselves feel bad because they want to feel good. What they want is fine, but the way they are going about it stinks. Rather than reframing all the other parts yourself, you can build a meta–part that recognizes this and says «Hey, you are doing this because you love this guy. Do you remember the first time you fell in love with him? Do you recall what that was like? The way you are trying to get him to treat you well isn't working. Do you remember what you did then? What else could you do? What does Janie do with her husband that works?» The meta–part goes into some way of generating alternatives: it provides ways for them to get what they really want. At specific times it says «Go in and change your behavior and get out of this loop; you've been here before and it has never worked. Arguing is not going to get you what you hope it will, and in order for you to argue, it must be really important. It must be important enough to change what you are doing.»

Man: I'm struck by that phrase «important enough.» I know it's important intuitively, and it has already been emphasized a couple of times, but can you explain what makes it so powerful?

It's a presupposition. It presupposes that this is more important than the other. If I go «Look, you are tall enough to reach that glass," the implication is that you are taller than I am. If I say to a kid «You are strong enough to stand up for yourself now," that presupposes that there was a time when he wasn't, but he is now. If I say «You are old enough to pay your own way," it presupposes that she wasn't at some time, but now she is and she hasn't recognized it. «You want this because it's important. And that means it must be important enough for you to do these other things.»

This is a great pattern to use in couple therapy. The couple is arguing and shouting «I'm right!» So you say «You're arguing because X, Y, and Z is very important to you. But is it important enough for you to consider other ways of communicating that might work better than arguing?» It's a great double–bind. If it's not really important, then they wouldn't argue in the first place, and they can stop now. If it is important, then it's important enough to try something else that might work, since what they're doing now isn't working. All the power behind being «right» gets channeled into new behavior.

In one couple I worked with, every time her husband would give her what she asked for, she would want more. She knew that she shouldn't, but she was dissatisfied and that made him crazy. He used to offer her things, but he didn't much anymore. She had a part whose primary intention was to get him to reassure her that he still loved her. What that part was doing wasn't working very well. I decided to build a part to help it: an ally. Any time she began to have doubts, this new part would come into action. This ally reframed the reassurance part on an ongoing basis. Whenever she had doubts, the ally said «Look, is it important to be reassured?» «Yes.» «Well, good. Is it important enough for you to find out what you can do to reassure him that you love him?»

This will result in a much broader behavioral change than simply giving her other ways to feel reassured. The ally will get her to do many things with her husband that don't get reassurance at the moment, but will result in her being reassured spontaneously at other times, which is what she really wants. You can't directly get somebody to reassure you spontaneously. But you can behave in ways that will eventually get it for you spontaneously at other times. A meta–part can be a good way of doing that.

Woman: I'm trying to relate this to the six–step model, which would say «Find out what need the present behavior that you don't like is satisfying, and find a new behavior that would better satisfy the need.»

Yes, that's the six–step model. One way to think about such «problems» is as if every behavior serves a need. Or you can assume that a problem behavior has nothing to do with their needs; it's just a byproduct of achieving some other outcome. That would also lead you to use the six–step model.

The difference between when the six–step model is particularly useful and when building parts is particularly useful is the difference between building parts that stop things and building parts that do things. With the six–step model you usually start out with some behavior you don't like and get new choices so that you no longer use the unwanted behavior. That's using reframing in order to stop something. The situations where building a part is most appropriate are those in which a person wants a part that does something: he wants to generate certain desired behaviors, and he is not doing it. When people ask for a part that stops something, then the six–step model, the secondary gain model, is going to be much more appropriate.

Man: How about building a part that differentiates between professional relationships and personal friendships? College professors who lecture at you when you are chatting could use a part like that.

Yeah, I can think of some people who could use that.

Man: How about a part that will give more flexibility to a person who has a lot of polarities?

Well, you have to be more specific about what you mean. You're being very general. What you're thinking of may be really groovy, but you have to be careful about how you describe it, because we've got this other human that we're going to install it in. Does this mean that he is going to become tolerant of having his parts fight with one another? What do you mean?

Man: Let's say the person has a polarity response to situations involving groups of people; you develop a part that will allow the flexibility for that person to listen.

Oh, you mean the ability not to have the polarity response. If you do that, you have to consider the possible secondary gain. If he always has a polarity response, is there some positive function? There may or may not be. The nice thing about your example is that if something is that overgeneralized in behavior, very often you can just build a part that listens to lectures and no other part in him will object to that, because there is no secondary gain to not listening in that context.

No matter what the difficulty is, you can act as if there is secondary gain and fix it. That will always work. If you pretend well enough, you can get anything to be real. But it may turn out that there is no secondary gain. There may only be secondary gain in a polarity response when you are being lectured by parents as a teenager. The polarity response allows you to rebel. However, you overgeneralized that response to all situations.

You become one of the people who sit at the back of one of our seminars. Afterwards, you say «Well, but what about insomniacs? It works well for phobias, but what do you do with depressives?» That person will leave the lecture not knowing anything about how to work with phobias because of his polarity response.

The point I'm making is that not listening may have no secondary gain in one context; it may in another. So if you just build a part for that particular context it may work great, but to avoid objections you have to be very specific about what it's going to do.

Man: A part that will get a person to show up on time for therapy sessions, or a part to do homework on time.

Which of the three models is going to be most appropriate for the example he just gave? «People are late.» What does that sound like? … It sounds like two parts tripping over each other's toes. So you'd use the negotiation model for that.

Woman: A part to discriminate between a dangerous situation and a safe one.

A part to discriminate between what's dangerous and what's safe. What do you think about that? What does that sound like? Does that sound like a situation in which you've got to 1) reframe one part, or 2) build a part, or 3) negotiate between parts?

Man: You could do any of those.

Well, you can always use any of these models, but which one sounds most appropriate? Woman: Build a new part.

Bill: Rebuild an old part. Take the part that has kept him safe enough to get here, that kept him from getting hit by cars or anything, and—

How do you know that? She didn't say anything about that. What happens if you have someone who's always stepping in front of trains? She didn't specify any of that.

Man: He must be getting missed by trains or he wouldn't be here.

That's a pretty big assumption. You can verify that with sensory experience, but I can think of examples of people who need to have parts that distinguish between situations which are dangerous and those which are not, because they get them mixed up.

Woman: That's particularly true of children.

Right. Your parents all built one of those in you. It's part of how you got here. Think of all the people who didn't make it to this seminar.

Let me back up and run through the whole thing again quickly. I made a statement at the beginning that one thing I noticed about therapy is that most of what people are doing is building parts. That's about eighty percent of what many therapists actually do. If that's true, then why is building parts so prevalent? Building parts is often inappropriate. I don't think everybody needs a «parent," a «child," and an «adult," but I think some people do. The question is «Who's going to need a part?» and then «What, specifically are they going to need?» What kinds of familiar contexts occur where people need parts? (Someone walks very noisily across the room.)

How about a part that gets someone to pay attention to sensory experience when they walk across the room, and to notice that they are making a tremendous amount of noise? We just had a demonstration of that need. That would be an exquisite part for some people to have. Perhaps in some situations lacking that part won't be detrimental. However, if you don't have a part that pays attention to how people are responding to you, there may be a lot of people who act as if they don't like your behavior, and you won't have any way of noticing that or changing that. There are many, many people in therapy who have that particular problem. They don't have any friends, and they don't deserve any. How many of you have had clients like that? You may tell them that somebody out there is going to like them, but deep down inside you don't like them. Often the problem is that they really have no way of knowing how people respond to them. That would be a really prime example of where it's appropriate to build a part. Where else are you going to need to build a part?

Woman: In couple relationships, you might need a part that would negotiate with your partner.

You get on the borderline there when you talk about having a part that carries on the negotiations. What are the rest of the parts doing in the meantime? I want to know what the outcome of installing this part would be.

Let me give you an example. A couple came to see me because they both had stupid behaviors that fired off automatically and prevented them from talking about what they wanted to discuss. I just picked one of them and installed an interrupter part. The new part did something that captured the attention of both of them, and interrupted the stupid behavior long enough that they could go back and talk about what they wanted to talk about. I don't know what you have in mind, but that's one thing that I've done.

Woman: You could install a part to tell reality from hallucination.

Now that would be a hell of a good part! Someday I may try building one of those. Later on this year I'm going to a state mental hospital to train the staff. This hospital's main function is to «warehouse» the patients that nobody really knows what to do with. A very interesting person has just taken over control of this hospital through an odd set of circumstances. The only thing that I intend to teach when I go back there is how to use this model to build a part that makes distinctions between what is shared reality and what isn't. Many psychotics do not have a part to do that.

Man: Many psychiatrists do not have a part to do that when working with those people.

Many do not have it at all, as far as I can tell! The only difference is that they have other psychiatrists who share their reality, so they at least have a shared reality. I make lots of jokes about the way humanistic psychologists treat each other when they get together. They have many social rituals that did not exist when I worked at the Rand Corporation. People at Rand didn't come into the office in the morning, hold each other's hands, and look meaningfully into each other's eyes for five and a half minutes.

When somebody at Rand sees somebody else do that, they go «Urrrhhh! Weird.» The people in humanistic psychology circles think the people at Rand are cold and insensitive and inhuman. Well, those are both psychotic realities, and I'm not sure which one is crazier. And if you start talking about shared realities, the people at Rand have more people to share theirs with.

You really have a choice only if you can go from one reality to another, and you can also have a perspective on the process of what's going on. It's really absurd that a humanistic psychologist who is hired to teach at Rand Corporation doesn't alter his behavior. Think about how often that happens. If someone who has all the TA jargon goes to a gestalt institute, he'll get wiped out by the gestalt therapist, because gestalt people can yell, and most TA people can't. In that context, as far as I'm concerned, the inability to adjust to the shared reality is a demonstration of psychosis. Qualitatively it's not different. You just don't get put away.

Let's say you're a gestalt therapist or a TA therapist or an analyst or whatever your gig is—even a Neuro–Linguistic Programmer. Tomorrow morning you wake up and go down to some bar where all the guys are. You say «You know, I was working on strategies this morning with a client, and I was watching his accessing cues — " They say «What?» So you say «Well, you know, I studied with Bandler and Grinder. We watch the movement of people's eyes and we know how they are thinking.» They say «Sure you do.» «Oh, yeah, and then I elicited a response and I touched him, and then I associated it with this other memory simply by touching him.» They would think you're pretty weird.

What you have to remember is that reality is contextual. Here we can talk about that and everyone nods meaningfully. You have all become a part of my psychotic reality. I have built a part in you so that I am no longer crazy.

If we had a psychotic, one of the things we could do is get the psychotic to have everybody else see his hallucination, and then we could call it «religion» or «politics.» The point is that realities are defined structures.

In some situations you need to build a part, and you can build a part in any situation. Who in here needs a part? Does anybody in here need a part? What do you want? What is the part going to do?

Teri: It would keep me at a certain weight.

You want a part to keep you at a certain weight. That certainly is a well–defined one. Who else?

Man: I want a part that enables me to see clearly with binocularity, without glasses.

It's a possibility.

Woman: I'd like a part that would be creative with metaphors.

I want you all to notice something. I'm now going to comment on group process. I asked people to define an outcome, and people were having lots of trouble. Just now I said «Who in here wants a part?» and listen to how well–defined the outcomes are! Remember that when you work with people.

Man: I want a part to remind me at regular and frequent intervals that I know what I'm doing most of the time.

All right. If you have to have one like that.

Jill: I want a part to let me know what I do know how to do.

What's the outcome of that? That's not an outcome. That's a process. This is the first slip–up we've had here. I want to know exactly what this part is supposed to do. You might get yourself a part to remind you of a catalog of your therapeutic skills, or to remind you to have variety in your behavior. That would be specific. What do you want?

Jill: I want a part that will move me on to something different after I'm convinced that I've already done one thing successfully.

OK. You could have a part that says «OK. You've done reframing; you know the standard model. You're hot at that and that would work here, but let's do it differently this time and have some fun.» Do you want that one? Or do you want a part that after you have successfully done reframing says «OK, it worked» so that you don't go ahead and use another model to cure the same problem that's already been cured?

Woman: I'll take both.

Man: I'd like a part that would allow me to relax when I'm sitting in a chair listening to a lecture, or when I'm talking.

Is that appropriate for this model? … It could be. The key question is «Are you relaxed at other times?»

Man: Yes.

So it's only when you come into a context like this that you get tense. Then the six–step model is more appropriate. If you built a new part to relax you, that would conflict with some other part that is making you tense.

Man: I'd like a part that permits me to retain the content of a lecture without taping the lecture.

Woman: I want an assertive part, but assertive only in certain contexts. Would I use this model, though, if I had an assertive part and the unassertive problem was contextually related? In other words, I have an assertive part… but it doesn't work all the time.

Well, I'm not willing to agree with that description. I would say what you are describing is that you have an assertive part, because you are capable of being assertive in some situations. This could be described in two ways. One is that you go into some specific situation and that Part goes «Not me!» The other possibility is that the assertive part says «Get 'em now!» and some other part goes «Shhhh!» The question is

«Which of those two possibilities occurs?» When you know that, then you can decide which kind of reframing to use—but either way, it's not building a part.

Harvey: I'd like a part to enable me to make lots of money.

OK. You want a greedy or clever part, depending upon how you think about it. Again, there are still lots of questions about your outcome. If the outcome is to get a lot of money, there have to be some well–formedness conditions about how you'll do this. Otherwise you might just take a gun and rob the first bank you see. This is the same thing I was talking about earlier. I want a specific description of what this part is going to do. Otherwise if I install this part, it may just go out and rob a bank.

Harvey: I want a part that will build referrals and find new markets for my skills.

OK, good. If you build the part that's going to generate referrals, then it knows how to go about getting money. That's a very specific kind of outcome.

Ray: I'd like a part that would enable me to improvise on the piano.

Woman: I'd like a part that allows me conscious access to visual images from the past.

You can't make eidetic images? How do you spell «greenwood»? … OK. You can make pictures; the emphasis is on the word «conscious.»

Woman: I'd like a part that would allow me to create hilarious humor whenever I wish. I want a part to just blow people apart into a humorous state.

Man: I want a part that will… allow me to … pause.

Do you want another one? You'd better be a little more specific, because you just demonstrated that you have a part that can allow you to pause. That's a mild incongruity in your communication, but my guess is that you have something more specific in mind.

Kit: I'd like a part to deal with «passive–aggressive» behaviors.

You will have to define that for me. «Passive–aggressive» is a double nominalization.

Kit: OK. I feel an incongruity in a set of behaviors that I experience with people and in myself—

We're going off into the «Land of Nominalization» here. We have to be careful. What is this part going to do?

Kit: Well, the part is going to serve a need—

All parts are going to serve a need. What is it going to do?

Kit: It will be like a periscope.

You can't specify outcome with that kind of metaphor. You've got to be very specific, or you'll go home at night and you'll lie down, and your body may come up like a periscope. If you say it's going to be a periscope, what exactly is it going to do for you? You've got to really tie it down to the world of experience.

Kit: I want balance between polarities.

Of?…

Man: You could build Kit a part to be more specific.

There you go. Beautiful. If you installed a part in her that was in charge of the Meta–Model, and then she Meta–Modeled her own internal dialogue, think what a gift that would be. Most of my students from the old days have a Meta–Model part.

Kit: I want a part to bring the stuff in me—

Wait a minute. Never mind. You need a part that knows the first four distinctions in the Meta–Model. In the exercise, someone is going to install a part in you that does that on the inside, so that before you speak, you can have the choice of using specific language or not.

Kit: I'd like a part that is … I feel that I… I've got a new statement to create it… since yesterday.

OK. Hold on for a few minutes. Go inside… .

Kit: I feel like I am.

Yes, that's right. This is called «pacing observable behavior.» Go inside and figure it out. I want you to make a visual image of what it is that would be occurring either in your mind or in the world of experience, that would allow me to know that this part that you want was operating. I don't want your image to be metaphorical. In other words, if you had a part that did this, what would I see that would be different? Take some time and do that on your own.

Lucy: I want a part to increase the frequency and intensity of orgasm with my husband. That would fit into this model.

True. Why not? Go for something worthwhile, I always say.

Woman: Is that building a new part, or is it just…

Well, the task of the part that she wants to build is not to make her have orgasms. Its job is to get another part to do it more often. The part she wants is what we call an «elbow part.» It gets other parts to go to work. It's like the difference between a motivation strategy and a learning strategy. One gets you to do it, and the other actually does the work. For instance, with a weight problem, it's one thing to build a part whose job is to diet, and it's another one to build a part whose job is to get you to go back on diets. There are different ways of thinking about how you are going to build parts and what you are going to build them for.

What we have been doing so far is step one: determining what specific outcome you want, and making sure that it is appropriate to build a part for that outcome. Before you build a part, I want you always to check to find out if another reframing model is more appropriate to get what you want….

For those of you who have chosen a part that you want and specified the outcome, I want to take you through step two. This is a fairly complex procedure, and we're going to do it methodically—not metaphorically. As a therapist you can dance around a little bit if you want to, but we're going to go through it step by step. I recommend highly that you have a piece of paper and a pencil during this procedure, because later there are going to be some things that you're going to have to keep track of, and there may be a lot of them.

I want all of you who are going to play this game with me to go inside and find out if you have any reference structures for the behavior that you want. Access any historical experiences that you can find of doing what you want to be able to do more methodically, more often, or whatever it is. Let's say somebody wanted to have a creative part. Perhaps they were only creative once ten years ago, or maybe they once thought of a new way to write a shopping list. Whatever it is, I want you to find whatever relevant examples exist. When you find those memories, I want you to step inside them, and get back all of the experience that goes with those memories. This is a very essential piece of what we are going to do. Take whatever amount of time you need to do that. Isolate specific instances of what you want, and be thorough about it. Make sure that you get at least one, two, or three experiences—if you have them—of what it is that you want this part to be able to do.

If what you want is a part that's going to make you have orgasms more often, don't remember an orgasm, remember a time when you got yourself to have one that you didn't expect. If you are going to build a part whose job is to get you to maintain a certain weight, you don't want to remember a time that you weighed that much, you want to remember a time when you maintained a specific weight.

Access those memories as intensely as possible. You don't need a long duration. The intensity of what you access will be the important part.

Woman: And if there are no examples?

If there are no examples, then wait. But be very careful, because there probably are examples of everything in your experience, whether you know it or not. If you can't find any examples, then just hold on, because step three is going to take care of that anyway… .

For those of you who could not find any examples and for those of you who are finished, I want to go on to step three, which has two parts. First I would like you to create for yourself a dissociated visual and auditory constructed image of how you would behave if you were actually demonstrating whatever it is that this part is going to have you do. So with the weight maintenance, you're not going to see yourself weighing a specific weight. You're going to see yourself engaging in the behaviors that would happen as you were at that weight, and as you gained some, and as you lost some—in other words, whatever behaviors would be operating as you maintained that weight. Most people are adept at losing and gaining weight. The problem is that when they get down, they go right back up. That new part's job is not to lose the weight, but to do things that result in maintaining the weight. You need to see yourself in that context: what you would look like, what you would be doing, and what you would sound like.

Next, when you see a whole sequence that you're satisfied with, step inside the image and go through the whole sequence again from the inside. Make sure that you like the feelings that you have as you do this, and do a good job. You will go through it twice: once from the outside, and once from the inside. First you visualize a good example of what you would look like and sound like doing the behavior. Then the second time through you find out what it would feel like from the inside. If you're satisfied after you have watched the whole movie from the outside, then step inside the movie and do it from the beginning all over again. If you're not satisfied with the feelings, go back and change the images as you look at them from the outside, and then go through it again from the inside… .

*****

Bill: When I go inside, there is an internal auditory which is not apparent to me when I'm looking at it from the outside. I don't know how to get rid of that. I can look at the image outside and see myself writing a manuscript. But as soon as I go inside, I'm sitting there at the typewriter and I'm starting to write out the first sentences, and a voice comes up that has all kinds of objections to what I'm doing. I don't believe that continuing that loop of going out and then back in is going to handle that.

OK, ask that auditory part if—for the purposes of your fantasy at this time—it would be quiet just for a few moments, because what you want to do is to find out if this is something you do want. If it turns out that it is something that you do want, you're not going to be able to get it unless this part agrees. That's the next step. So if it has an objection to your having the internal experience that you fantasized that you want, then it will have a chance to object, and rightfully so. But first it should allow you the opportunity of finding out if you want the new behavior. When it allows you to have that experience, you may discover that if you don't have that internal voice, you have nothing to write. If so, you would then have to construct another fantasy.

Woman: Are you asking for one fantasy or several fantasies?

You just want one fantasy which is an example of the way you would look from the outside, behaving as if you had this part. And then if you like it from the outside and it looks safe—no one strikes you, you don't fall off a cliff or anything—then go through the same sequence from the inside, and find out if you like the internal experience as well. Sometimes you think that you want a part, but when you try it out, you don't like it.

If you're not done, just go ahead and take the time you need to finish the third step. It will do you no good to jump ahead. I'm going to go ahead and give the instructions for the next step. If you miss them, don't worry. I'll probably end up giving them several times, because this step is a little complicated. Make sure you finish the step that you're on first. Take all the time you need. I'll go over it as many times as you need.

When you have the well–formed example of how you would behave if you actually had this part, and you are satisfied with that fantasy from the inside as well as from the outside, then the next thing I want you to do is an ecological check. This is step four. I want you to go inside and ask—and it is very important how you do this — «Does any part object to my having apart which will be in charge of making that fantasy a reality?» That is a yes or no question. If you get a verbal «yes," fine. If you get a feeling, you have it intensify for «yes» and diminish for «no.» You can use all the methods from the other refram–ing models for this.

If a part objects, I want you to ask «What is your function for me?» This time you don't care what the objection is. That's not the important part. You want to find out what it is that the objecting part does, what its job is, its function.

When you get that information, if it doesn't make sense to you that it would object to your having the new part, go ahead and ask it what, specifically, its objection is. Ask it how it thinks this new part is going to get in its way.

Let's say you decide «I'm going to install a part that teaches me how to hold my breath for an hour and a half.» Then you go inside and ask the question «Does any part object?» You get a «yes» so then you ask «What is your function for me? What do you do for me? What's your job?» The objecting part says «Well, I'm the part that keeps your heart beating.» If you can't consciously tie together how holding your breath for an hour and a half is going to interfere with the heart–beating part, then I suggest you ask. My guess is most of those connections will be obvious. But if they aren't, then ask.

The pencil and paper are for this step. So far each time I've used this model there have been at least eight or nine parts that object. Depending upon what part you are building, some of you may not have very many objections. When you go about installing a new part, the potential for it to get in another part's way is a lot greater than when you're just altering one part's behavior a little bit. There may be many, many parts that object to creating a new part. The more the merrier, because they are all going to become allies in the design process. Just make a complete list of the parts that object, and each one's function. You want to know if there are parts that object, and if so, what do they do? What's their job? Be thorough. Make sure you get all of them. Check each representational system for objections, to find out about all the Parts that object in any way. The objections will be the essence of making sure that the part you build is really graceful and works well. They will be the talents of the part that you are going to build… .

Lucy: I got six parts down.

OK, you've got six parts down. And you got all of their functions?

Lucy: Oh, no, I've just got the parts listed.

OK, I want you to find out what each one's function is. You say «Part number one, you objected? What do you do for me?» You don't ask «What is your objection?»; you ask «What's your function?» You want to know what each objecting part is in charge of. It's not that you have a part that says «All right, we're going to limit the orgasms here.» It's just that the objecting part is doing something else. When it considers the possibility that you would have a part that gets you to have orgasms more often, it says «Hey, I'm not so sure I want a part that does that.» Now, you might be able to install the new part anyway, and perhaps the part that objected wouldn't interfere with it at all. However, if we find out what its concern is, then we can build an even better part, and be sure that other parts won't object to it.

Lucy: So you want me to find out what each part's concern is?

Secondly. First I want you to know what its function is, what it's in charge of. If that doesn't give you an understanding of what its concern is about having this new part, then ask.

Lucy: I'm not sure if I understand. For instance, there is a part that doesn't want me to put pressure on my husband, and a part that doesn't want to give me what I want. Now would those be two parts?

OK, now, what's the function of the part that «doesn't want to give you what you want?» I'm sure that you don't have a part that just sits around and says «What can I keep from Lucy today?» It's got to be in charge of some other task. The question is «What is its job?» It may be a part that doesn't want you to have unrealistic expectations. All I want to know is what its function is, and what its concern is. If you do have a part that doesn't want you to have «unrealistic expectations," then you already know what its concern is. Its concern is that the new part isn't going to work and that you'll be disappointed. You don't have to worry about that one, because what we build will work. You want to know the function of each part that objects, and a little bit about what concerns it about having this new part around.

This new part that you want to build is going to influence your behavior. You want to know if there are other parts of you that object to it having existence in you. We want to know about every part that objects to the idea of having this new part. And we also want to know specifically what it is about having this part that concerns the other parts. That is very, very valuable information. We need to know that so that when we build the part, we can build one that is going to be satisfying to the total person, rather than just ramming something in and letting conflicts evolve. There are usually plenty of conflicts already; we don't need to build in more.

I've said nothing yet about actually installing these parts. So far we're just designing. The fifth step is what we call «satisfying well–formedness conditions.» The well–formedness condition of our design is going to be that no other parts object. We are going to take all of their concerns into account and modify the new part accordingly. We don't want to step on anybody's toes here except the conscious mind's. It's the only one who deserves it.

The fantasy that you had last time is the basis upon which those parts made their objections. You made up a fantasy, and a certain part went «Ugghh, boy, that's going to be hard.» Another part said «I don't want that!» Some other part said «If we do that, we won't be able to do this.» All the parts that had concerns based their objections on that fantasy. So now you're going to make a new fantasy. We now have a list of well–formedness conditions to use to modify the last fantasy and take into account all of the concerns of those other parts. Before you build the new fantasy I would like you to redefine your part so that it takes into account all of those concerns. This is the importance of the amount of time I spent on definition. For example, what were the functions of some of the parts that objected?

Teri: There was a part of me that said that if I maintained the weight I wanted, I might not be a therapist. I wouldn't want to do therapy; I'd want to be outside doing other things.

That was an objection?

Teri: Well, if I stay overweight, I'm comfortable doing what I do now, because I don't feel like doing much outside. But what is that part's function? Teri: To keep me the way I am.

No, that's not its function. If there's a part that says «Look, if you lose all this weight and maintain it, you're not going to want to do therapy," then you say «Well, OK, that's a possibility. What's your function?» If it says «Well, my function is to keep you the same» that's called «jive.» In this group we know that that's not a function, that's a behavior. What we're looking for is a positive function. You don't have a part whose function it is to keep you the same. You have a part that Wants to keep you the same so that X, Y and Z won't happen, or will happen. If you don't get functions, you're not going to be able to come up with a good set of well–formedness conditions.

Teri: For example, I wouldn't have much time alone with my husband.

OK, so there's a part whose function is to make sure that you spend time with your husband. And one way of carrying out that function is to do therapy, so you can hang out with him at work.

Teri: Right.

Now, that's a concern that makes sense. However, there are lots of ways to build in spending time with your husband. So the part of you that is going to maintain your weight is going to make sure that this part is very satisfied. It will build spending lots of time with your husband into the maintenance program. This now becomes part of the fantasy that you have to build. You don't become thin and run off, because this part isn't going to go for that.

OK, now I want another example of a function.

Pat: The function of my new part is to let me know when I know something. The concern is that it will defeat my motivation to learn more.

OK, so if you had a part that let you know when you know something, then you would be less motivated to learn. So now you have to build a desire to know what's over the next horizon into the new part that you are going to build. That's got to be an integral part of having a sense of what you know. «I really know reframing; I wonder what else there is to learn?» That's got to be built into the new fantasy part.

For each objection, you're going to have to modify the fantasy until it satisfies each of those conditions.

Bill: I've got one, the function of which is to keep me honest.

That is not a function. That is a behavior. That is an example of what we do not want. What is it trying to do by keeping you honest?

Bill: Protect me from being accused of lying … or to be honored if a person tells me I'm honest.

Well, make sure which it wants. There's a real difference between being honored and—

Bill: It feels more aversive than it does—

All right, but be persnickety. Go in, set up a yes/no signal, and find out which it is. Check it out and be thorough.

Teri: I came up with the main thing behind all this. The only time that I've ever been thin, I was crazy. There is a part that's not willing to let me become thin, because I don't want to be crazy.

Well, I certainly think that you could build strict controls on your mental health into a part that was going to put you on a weight maintenance program. Think about it this way, Teri. Poundage and sanity have no relationship to one another, other than anchoring. Whatever weight you weighed when you were whatever you are calling «crazy» had nothing intrinsically to do with being crazy. That time it was a coincidence. There's no causal relationship between losing weight and your being crazy again.

It's the same as when we did the lying exercise last week. I said «OK, it's time to lie.» Everybody else said «Yay, we're going to lie!» but you said «If I lie, I'll be crazy. I lied before and I was crazy, so if I lie now and I can't tell the difference, I'll be crazy, because that's how I was crazy before. If you want to make them crazy, that's all right, but I'm going outside!» … And the rest of the people were muttering «Well, we have to stay in here and be crazy.»

When I came outside I said to you «Look, lying, and not knowing whether you were lying or not, was an anchor for being crazy for you. The rest of the people in the room haven't done that—as far as they know. When you do it this time, we're going to give you a new anchor. We're going to make some black lines around your pictures when you lie, and they are really important black lines. Now you can lie, and know you are lying, so you won't be crazy this time.»

All I really did was to put a literal meta–reframe around what was going on. It's a very small change in your strategy. You're still making constructed visual images. We're just separating the constructed images from the others with those black lines, and it's that separation that allowed you to have new feelings about what you were doing. When I said «OK, now make up a lie» you made up a lie. I asked «Are you crazy?» and you said «No.» I asked «Which one is the lie?» «Well, the one with black lines around it is a lie. Anybody can see that.»

The point is this: if you are going to lose weight and go back to a weight that is associated with being crazy, you also have to put a new frame around it. That has got to be a function of this new part. If you are going to go back to the weight you had when you were crazy, then of course there's the implication that you are going to be crazy again. However, since you are not the person who went crazy, you can't possibly go back.

That distinction has got to be built into the part that you are building now. It's very essential that it's capable of distinguishing all the differences between this Teri and the one that went nuts. This Teri is a different age, and has a lot more information about the world of experience. You know how to do a lot of things that the other one didn't know how to do before. That Teri didn't know anything about NLP. She didn't know how to anchor psychiatrists. There are lots of tools that you have explicitly now, that you didn't have then—not only the things you've learned here, but in TA, and lots and lots of things that have nothing to do with psychotherapy. So this new part has got to be able to point out, now and then, how you are different.

Are there any other questions about how you utilize objections to redesign the part you want to build? This is a very important part of this process. This is the crux of building parts that will be functional, so that you don't end up being an anorexic or something like that.

Bill: I want a part to allow me to write papers comfortably, without agonizing. There's a part that objects. When I ask «Well, what's your function?» it says «My function is to make you aware of all the possible objections to, or criticisms of, what you are writing.»

OK. The contradiction I'm hearing is that that part wants you to write. Its job is to help you with your writing. Is it objecting to your being comfortable?

Bill: I like to write; the problem is that I sweat blood doing it.

That's impossible. That's the same contradiction. You're saying «I love to write; it's so painful.»

Bill: A friend of mine once said «I don't like to write; I like to have written.»

Well, that's different. — Bill: What I'm aware of is that I sit down at the typewriter all fired up. «Gee, this is some great stuff I'm going to put out here.» And then I go «Ughhh!»

Yeah, I agree. And do you know what you need to do in order to satisfy this part? One simple solution might be to ask it to read your papers at the end of each page.

Bill: Well, I was considering asking it to save its objections until I have a finished first draft. There's this other part that says «When you sit down, you write the first draft.»

OK, but you don't need to deal with the specific objections. You want to find solutions that preserve the functions of the parts that object. The important thing is that it objects. It's saying «Look, if you write this stuff, you're not going to consider all the objections people would have.» What if you had made a fantasy where you sat down and wrote comfortably, and then at the end picked up your work and let this part go to work with a red pencil to revise it? Would it be satisfied then? This is what I do with other people's writing. I have a part that does that really well, especially if it's somebody else's work. They call that «editorial work.» That can be a very valuable part.

Lucy: I'm hung up on a part that doesn't want me to give up control—

Of … ?

Lucy: —which I've taken to the part that will— Wait a minute! These forms are very important. You have a part that objects, right? What is its function? Lucy: Its function is to keep me from feeling helpless. That is not a function; that is a behavior. Lucy: It protects me from hurt.

That is a behavior. That is not a function. What will happen if you feel helpless? What is it trying to do for you by keeping you from feeling helpless? So what if you feel helpless?

Lucy: It's a needless function.

No, it's not. It's just that you don't know what it is yet. What is it trying to do for you by protecting you from feeling helpless? That's the first part. The second part is how does having orgasms more frequently have anything to do with feeling helpless or not? I don't need a verbal response to that. The point is to get the function clear so that you can understand how you have to modify your fantasy so that that part can be satisfied as well.

Ray: I was very surprised that any part objected to my improvising on the piano. The first objection was that I may end up playing the piano more at gatherings and interacting with people less.

OK. I didn't ask you to ask that question. This is important. I said if there is a part that objects, do not ask it what its objection is. Ask it what its function is, so you know what that part is in charge of doing. Then if you don't understand how it would end up being concerned, ask it what it's concerned about.

Ray: And I asked it the other way around.

Right. It is very important that when you're in my group you ask it my way. When you go home you can build parts your own way. If you don't know what an objecting part's function is, it's very, very hard to Please it.

Bill: I asked the question «What do you think will happen if you get critical feedback?» And the answer that I got was «If I get that, then I'll feel lonely and inadequate.» So it's keeping me from feeling lonely, and inadequate.

Well, in one sense that's the same answer. It's just a rephrasing. «Well, you know, it's my job to keep you from feeling lonely and inadequate.» «Well, what would happen if I felt lonely and inadequate?» «Well, you would feel bad. My job is to keep you from feeling bad.» «Well, what would happen if I felt bad?» «Well, if you felt bad, then you would feel like people don't like you.» What it's doing is just redefining the same thing that it doesn't want: for you to feel bad. But the redefinition does give you some more information. One of the things that you can do is to build into the construction of the part either protection against people criticizing you, or a way of enjoying criticism. You could build in the understanding that when people are criticizing you, it offers you the unprecedented opportunity to do a lot of things. One is to demonstrate how loudly you can yell. Another is to utilize their behavior. When people are angry at you, it's an unprecedented opportunity to test your utilization skills. There are so many opportunities out there. So you don't have to feel lonely; you can make criticism the basis for a lasting relationship.

I'll tell you one of the odd things that I noticed growing up. Everybody that beat me up became my friend, and vice versa. Growing up as a teenager in a very rough place, I discovered that one of the best ways to make a permanent, lasting friendship was to beat the snot out of somebody. I became conscious of this sometime around the eighth grade. If there was somebody I wanted to hang out with, one of the fastest ways to become friends was to go beat the snot out of him. I don't know how that works, but it's an interesting phenomenon.

Man: Haven't you found some contexts where that behavior is not appropriate?

No, I haven't yet, actually. My whole professional relationship is built on it. I go around the country insulting people, and they pay me money. It's weird!

Harvey: I want to give you some more of my list of functions to see if I understand this. To keep me from failing is one. Giving me play time is another. Those are functions, aren't they?

Sure.

Harvey: To get love from others.

To elicit a particular response, yes. Hopefully that part knows what it means by the nominalization «love.» You might have it be a little more specific. That's important.

Harvey: OK. I've got another one: To be a caring person.

That is a behavior, not a function. I'm going to teach all of you this information yet. You may not learn to build parts, but you're going to learn what a function is by the time you leave here! «To be» is a description. Listen to the phrase «to be or not to be.» When you make a statement»… to be something or other» it's a description of a behavior. «… to be angry. I want a part to make me be angry.» That isn't a function; that's just a description of a behavior. The part wants you to care about what or whom, where and why, and for what purpose? What is it going to get by having you be caring? What would happen if you weren't?

Harvey: The idea behind it is that I don't want to become like a machine.

That's the ultimate in humanistic psychology. «I don't want to be an android; therefore I'm going to act like this all the time. I'll hug everyone; that way I will not be a robot.» The point is that this part wants you to do something that constitutes «caring.» I have no idea what that means. Does that mean that you tell people honestly what they really ought to hear? Or does that mean that you touch everyone? What does it mean? Don't answer, because I don't care what the content is. I want you to know, and I want you to know what the function of that specific behavior is. It may be that that part wants you not to be an android, in which case all of this has nothing to do with being caring or not. You may go back inside and say «Have you ever thought about how much being 'caring' can be androidal? Let's have caring and not–caring; we'll alternate days. At least I'll be a different android.»

The question is «What is the part's function?» If the part's function is to keep you from being an android, then the question is «What does that mean?» Does it mean not having repetition in your behavior? Does it mean not doing all the things that Maslow said were bad? It's essential to find out what the part's function is.

Ray: This time I asked the question about function and the response was «I'm here to take care of you so you won't become like your father.»

Only a psychologist would say that. Ray: My father improvises on the piano.

Right. But that still isn't a function. You have to go a little bit further than that. If you were to be like your father was, improvising on the piano, then you would be what? There was something about the way your father improvised that some part of you thinks is negative in some fashion. He either made a fool of himself, or he did something else that some part of you didn't like, right? Now, what was that? Ray: He avoided interacting with people.

OK, so there's a part that wants you to have personal interaction with people. Good. Now all you have to do is build some way to have personal relationships with people into the way that you improvise on the piano. You have to define what «personal» means, because obviously it isn't singing songs. Maybe it means that you have to be able to play background music and have meaningful psychological conversations.

Woman: Is «to be taken care of a function or a behavior?

That is neither. It's so unspecified, it's nothing. «To be taken care of—how, specifically, … in what way? … by whom? …

Woman: Way back then is what I'm after, when I was in my original family.

You're going to build this part that's going to do something, right? What is it going to do?

Woman: It's going to let me be comfortable dancing in front of groups of people.

Let you dance? You've used the wrong reframing model, because there's obviously a part that stops you from doing this.

Woman: Yeah, I know. I knew that to begin with and—

But you thought you'd slide it in anyway. The point is that obviously there's going to be a part that isn't going to like this, because its job is to keep you from dancing in front of people. That part will have a very strong objection. So there's a part of you that objects, and its function is to have you «be taken care of.» Go inside and tell it «I don't have any idea what 'to be taken care of means. What specifically does that mean in experience?'

Woman: In my original family I had to do X, Y and Z to be taken care of.

OK. You had to take out the garbage in order to get a Twinkie, but I'm asking «What does that mean in your experience now?» What you've got now is so unspecified that I can't help you. It's like looking at somebody and saying «Noun, verb, adjective, noun.» There's no content in your sentence, so I can't even respond to it. It's just a little too formal. What you want to know from this part is «What do you do for me as a person now? " You have to go back to it and say «Look, I need to know in experiential terms what it is that you do for me as a person. You are obviously a part of me. I do not live with my parents anymore. I want to know what it is that you do for me, and how you are concerned about my dancing. If I dance around, what's going to happen that's so bad?»

You want both those questions answered. You want to know what that part's job is. «What do you do for me?» «I get you to be taken care of.» «How do you do it? What is it that you do?» You see, «being taken care of might mean that people hug you. Or it might mean that people feed you. Or perhaps it means that people are nice to you. You need to find out what that means in the land of experience. That is what counts in the end.

Dan: I want a part to fully access visual and auditory information that I take in. A part objects to reaccessing totally all visual and auditory information.

And I agree with it. What's its function, though? What does it do for you? What would happen if Dan actually had a part that could recall all visual and auditory information? People like that are called idiot savants, and they get put into mental hospitals. Idiot savants are completely dysfunctional; they can't operate at all as human beings. They are constantly aware of everything that ever happened to them. They can multiply great. They are whizzes at mathematics, but they can't function in the world of experience, because they have so much internal «downtime.»

When a part has a concern, take it into account and then redefine what the new part is going to do. Dan might redefine it as «I want to build a part that is going to make available some information about all the anchors that are occurring in a particular environment," or «I want to be able to visualize specific pages in a book I read once.» Whatever it is that you want to be able to do, specify it in experiential terms so that you know exactly what is going to happen, and so that the other parts of you know whether they are going to object or not. Then if they have objections, they'll be good ones.

Bill: I've got a function that I'm having a lot of trouble phrasing. The stuff that I'm getting says «Look, I want you to minimize the probability of being laughed at and treated with disdain. And I want to maximize the probability that you will at least be treated with respect, and hopefully even honored.»

OK. Well, that certainly is a rather extensive function. It doesn't Want you to make a fool of yourself.

Bill: OK. Is that a function?

Yeah, that's a function. However, the question is what would happen if you made a fool of yourself?

Bill: That's not my question. My question is «Are those words which convey function as opposed to words which convey behavior?'

Well, those words mainly convey behavior, but the functional implications are tied to the behavior pretty well.

Bill: What are some words that do convey functions?

OK. For example: «If I made a fool of myself, then I would lose income because my clients would go away.» The part's behavior is protecting you from making a fool of yourself because it's afraid that you will lose your friends and your business associates and your clients. The function is to keep your friends.

Bill: So it's the «then» part of an «if, … then» statement?

That's one way to think about it. There are many ways of getting the function. It's just not as simple as you want it to be. The function is what you get; behavior is how you get it.

Lucy: I'm on my list again. A part that rationalizes objected and then said «It doesn't matter.»

Ask it if its objection is really important… .

Lucy: Nah. I've got another part that doesn't want to give me what I want, because I've already got so much it isn't fair. That is its reason.

I didn't ask you to find out reasons. What is the function of your having a part like that? That's the important question. We don't want to know the part's reasons.

Lucy: OK. I'll have to look at it again.

What kind of part is it?

Lucy: A bad part; I don't want it.

No, No, NO! There are no bad parts. We have all good parts. The only question is, " What kind of good are they doing?»

Lucy: It's making sure that I deserve the things that I get.

And rightfully so. OK. So that's its function, and its concern is that if you get this too easily, you won't appreciate it. Then you build in having to do certain things that make you deserving of the outcome you want to have.

Nancy: I can take care of all the objections and concerns, but then I'm getting a message that I need another new part—an ally—to help do all that.

Fine. However, I think it probably would be better if you found an ally that's already in there. You don't need to build a whole new ally–Just go in and find some other part that can already do what you need.

It's too much work to build two new parts in the same day.

Nancy: One part's concern is to be sure that I would be able to notice other people's disapproval.

What is the function of the part that wants you to do this?

Nancy: Well, it would be the part that is concerned with my failing.

It's not concerned with whether you fail or not; it's concerned about whether people disapprove of you or not.

Nancy: Well, yeah. The two are connected.

Certainly. If you fail, people may disapprove of you. But what it's concerned with is people's opinion of you.

Nancy: Yeah, and it's also related to getting some things that I want.

OK. That part is concerned that if they disapprove of you, then they won't give you a job, or they won't give you something else. Are you telling me that you don't believe you have the sensory experience to tell when you are behaving in a way that people disapprove of? I want a literal answer to that question. Do you, or do you not, have that much sensory experience? …

Nancy: I guess I do.

OK. Just go and get the part that does that, and say «Hey, part, I need your help over here. We're building George, a part who's going to do X, but George needs to have sensory experience about this, and you know how to have sensory experience. I want you to connect with this new part if we decide to build it.» The new part doesn't have to have every quality. If you already have parts with certain qualities, you just get those parts to help out.

When you have all the functions and concerns of all the parts that object, then go on to redesign the fantasy so that all those concerns are satisfied. Check with every single part to make sure that each one is satisfied that this is a representation of an outcome that includes all the things they are concerned about and whatever it is that you want the new part to do. You make a new fantasy in which all the objections don't need to exist.

So in Teri's case the new fantasy must include the new part having responsibility for making sure she distinguishes between the way she is now and the way she was when she was «crazy» so that she can tell the difference. As long as she can tell the difference, the part that doesn't want her to go crazy won't object.

If there's a part that's afraid you are going to make a fool of yourself, then you have to build in the ability to have the sensory experience to notice when that happens. Without that you won't know, and those other parts are always going to be afraid that if you do certain things, you'll make a fool of yourself. You have to have feedback loops to satisfy those parts. If Bill writes, then he's got to be able to have ways of becoming somebody else and reading his writing in order to find out whether his response is «That's stupid!» or «Hey, that's pretty interesting.»

Whatever concerns your parts have, you build safeguards into the fantasy so that you have a representation of what the world of experience would be like if you had a new part that functioned to get the outcome that includes all the concerns the other parts have. We not only want to build a good part, we want to build a graceful part as well, a part that can do what it does and not step on any other part's toes.

Now, if you have all the objections and outcomes, go ahead and build the new fantasy. First do it dissociated, and then do the whole thing again from the inside. Revise that fantasy until no part objects… .

If it's possible for you to make up a fantasy of yourself doing something and then step inside it and have the experience of doing it, you have already made all of the necessary adjustments at the unconscious level—in terms of strategies, in terms of representational systems, in terms of everything relevant—for that part to know how to generate the behavior and behave in accordance with what you want to accomplish.

A music teacher once told me «If you can hear something inside, then you can play it outside. The trouble is learning how true that is.» Another person that I took music lessons from said «When you can hear it on the inside, then do it with your mouth. And if you can do it with your mouth, then you can do it with your hands.» What was behind that is the understanding that in order to make a fantasy, you have to do everything mental that needs to be done in order to actually do it in experience. For instance, in a detailed fantasy you see somebody and tell him a joke and he laughs. As you did that, you had to go through all the necessary strategies to develop the joke: the creativity, the gestures, the talking—all of those behaviors are functioning in a detailed fantasy.

Step six is much more metaphorical than we usually are, and it demands some congruence on your part. You ask your unconscious resources to analyze that fantasy and to pull from it the essential ingredients. What you want your unconscious to do is something it does all the time anyway. You want it to take those first memories that you started with—the times in your past when you had actually done the behavior or something similar—and the fantasy that worked, and all the underlying structures, and you want it to give all that entity. That is how you randomly got all your parts anyway. Your unconscious can do that for you; it does it all the time. Those of you who are TA therapists and have TA parts, who do you think made them? You made them.

So you go inside and you say «Look, either one of two things is the case: either my unconscious mind is in charge of building parts, or I have some part in there that builds parts. Whichever it is, I want you to build this one and to give it entity. You don't do this step until you have a well–formed fantasy that no part objects to. When you've modified the fantasy so that every part is satisfied, then you do this step. You have your unconscious or some part of you give it entity so that it will function on its own.

Man: Could you go over that again?

You want to get some part of you, or your unconscious, however you think about it, to analyze the fantasy. What I'm saying is «Look, in the fantasy you used a strategy that worked. You want the part that you are building to use the experience of that fantasy as a foundation, and to operate out of that strategy.» This is not what you would tell a client, by the way. You use whatever metaphors you need to with a client. Whichever way you think about it, whether you call it a strategy or a part, you say «Look, go in there and get what you need to know.» This is what I would tell a client. «Get what you need to know from that fantasy to be able to build a part of you that can do this exquisitely and easily, and at every moment that it needs to be done.»

I want you to go ahead and run through the rest of the steps. If you get stuck, let me know. Then I want to demonstrate to you some ways of testing… .

Are you done with that step? OK, now we get to the most important step of all: step seven. We have to test the part to make sure it's there. There are several things you can do. You can go inside and ask «Are you in there?» That's always a good first step. You can also do things behaviorally that would engage that part and find out what happens. You should also add lots and lots of future–pacing activities to your testing phase, especially when you do it with other people.

Now, how is your part going to work? Tell me something about what it would do. When would it be apt to do something?

Teri: Well, if I were to get on the scales and weigh two pounds more than what I want to weigh, it would let me know that I need to stop eating so that I can get back down.

So it would put you on a diet. Will it design the diet for you, or is there another part that does the dieting?

Teri: I already have a part that does that.

OK. You have another part for that. This part's job is to say «The time is now.» What else would it do?

Teri: When I get to the weight that I want, it will give me a picture different than what I looked like when I was crazy.

What would it be apt to do right here?

Teri: Ah, it would allow me to go into the dining room and be with people while they are having dinner, and only eat what I really want to eat, and not allow me to eat everything.

It will disconnect the automatic hand–arm–mouth sequence? OK. How would I know if that part were active, Teri? If I walked into the dining room, what would I see?

Teri: You'd probably see some food left on my plate, depending upon whether it's something that I like or not. Visually, I'm not too sure that it would show for a while.

Is this part in a hurry to lose weight or does it do it slowly?

Teri: Slowly. I'd only agree to doing it slowly. I've already worked it out with my husband. If he wants to eat, he can eat, and I don't have to.

All right. Who else is done?

Bill: My part is going to wait until at least—

Shhhh! (with disdain) I didn't call on you… . Well, what happened? I was just testing the part. You built a part to help you respond to criticism, right?

Bill: Well, I'm recalling what… I was amused … and responded in a way that … helped me.

As opposed to?

Bill: Feeling that I have just done something awful, and I should be embarrassed, or ashamed of myself, or coming back with an angry response.

OK. Anybody else?

Pat: I have a new part, and I even made it grow, so that it is now about twenty–five years old.

Ah. I like that. OK, you built a part. What's it going to do?

Pat: It lets me know what I know, and to act on it when I know it. And my test is now to tell you that I have this part, and I also know that I know—

Give me an example of an outcome. If you were going to build another part, what would be an example of a function that—

Pat: I am going back home and I am going to use it everywhere. I am going to teach it; I'm going to—

That's what this one does. Now, give me another one. If you were going to build another part, what function would it have? If I were going to bring you up and have you demonstrate building a whole new part in another area of your life, what would you build? What would its function be?

Pat: Its function would be to utilize my unconscious mind as much as it can, as it did just now.

OK. Now, is that a function or is that a behavior?

Pat: Its function is to extract everything of value from my unconscious.

OK. Is that a function or a behavior?

Pat: I think it's a function.

OK. Are you sure, though?

Pat: It has to do it in action.

Yeah. I'm just asking you if you are sure, if you want to commit yourself to this. Pat: I am sure. It's a function.

OK. Are you all learning how to test these things, by the way? … Anybody else? Man: I think I got it.

What was the part you built? What does it do?

Man: It's a part that says «Go out and try something that you think you know. Try to do it, and see what happens.»

So that part will get you to do it, to find out whether you really know it or not. That sounds like a good idea. That's what you built the part for, and you finished, right? OK, what do you think you know but you're not sure if you know it?

Man: Just now I wasn't sure I could tell you that I thought I had succeeded in building that part without getting in trouble.

Did you?

Man: Yes.

OK. Well, what else did you learn to do that you are not quite sure you can do?

Man: I learned to elicit a strategy completely enough so that I know that one step of the strategy moves logically to the next step.

Good. I'd like you to teach a group about that tonight for people who don't understand it, here in this room at 7:30.

Man: All right.

OK. Anybody else?

Man: I have a part that allows me to speak in a group situation. This is the first time I've heard from him! How did it feel to talk? Man: Fine.

I recommend that each of you try building a part for someone else at least once before you leave this seminar, to find out what happens. There are a lot of people in the seminar who weren't here in this session. Go and test this model on somebody during lunch and find out how it works.