4) Access a creative part, and generate new behaviors to accomplish the positive function.

a) Access experiences of creativity and anchor them, or ask «Are you aware of a creative part of yourself?»

b) Have the part that runs X communicate its positive function to the creative part, allow the creative part to generate more choices to accomplish that function, and have the part that used to run X select three choices that are at least as good or better than X. Have it give a «yes» signal each time it selects such an alternative.

5) Ask the part «Are you willing to take responsibility for using the three new alternatives in the appropriate context?» This provides a future–pace. In addition you can ask the part at the unconscious level to identify the sensory cues that will trigger the new choices, and to experience fully what it's like to have those sensory cues effortlessly and automatically bring on one of the new choices.

6) Ecological Check. «Is there any part of me that objects to any of the three new alternatives?» If there is a «yes» response, recycle to step 2 above.

Whatever «obstacles» you have encountered in doing reframing, I want you to select three that you'd really like to have more choices in dealing with. Then I want you to do an exercise in groups of three. Person A is going to look at his or her list of «obstacles» and role–play one of them as a client. B will then role–play an NLP programmer and try out ways to cope with the situation. Person C will be a consultant to keep B from falling into content and to keep B oriented.

For instance, if you are A, you will say something like «You've established rapport with me and set up a signal system with the part that runs X. We're on step three: you just asked the part if it will communicate its positive intention to me consciously. The response I've gotten is that I don't experience the old signal at all, but I have two different signals.» So A will set the stage at exactly the point in reframing where A wants more choices.

B will then try out one method of responding to the situation that wight move A toward the next step of reframing. C will be an observer, or meta–person, and notice whether B's maneuver is effective or not. Then I want C to ask B to think of two other responses to make to that situation, and then try out each of them.

Let me give you an example of the way I'd like to have you do this exercise. Let's say Beth is going to play client, Scott is going to play programmer, and Irv, you are going to be the meta–person, the consultant. Part of your job, Irv, is to observe and listen to the relationship between Scott's behavior and Beth's. At any point in time, I should be able to walk up to you and say «Tell me something about the relationship between the programmer's tonality and the client's tonality» or «Where are they in the reframing format?» So your job is to know everything that's going on—which is impossible, so just do your best.

The second thing that Irv is responsible for as meta–person is more specific. Any time the programmer hesitates or begins to be confused, you interrupt and say «Hold on. Which step of reframing are you on?» «Step two.» «What specific outcome are you attempting to get? What's the next small chunk of outcome you are going to get?»

Scott should be able to respond specifically, for example: «I want to establish a robust involuntary unconscious signal system with the part responsible for the behavior.» Then Irv will say «How, specifically, are you going to do that?» Scott will respond «I'm going to access it behaviorally by pretending to do behavior X myself and thereby induce it in her. Or I can ask her to do behavior X. Or I could ask her to go inside and ask the part if it will communicate, and make sure that the signal system that comes back is involuntary.»

Every time the meta–person interrupts, I want him to get not just one choice, but three options for proceeding. First you'll find out what specific outcome the programmer is going for, and then you'll get three ways he can attain it. These ways won't necessarily all work, but building in at least three options at every choice point will make you much more effective in your work. If you've got only one choice, you are a robot. If you've got only two, you are in a dilemma. However, if you've got three, you begin to have behavioral flexibility.

This is what I was asking you to do earlier when I was role–playing a client. You got access to the part in one way, so I said «Now go back to the choice point and do it some other way.»

Meta–person, the third thing I want you to do is to interrupt if you don't understand what's going on. If Scott is the programmer, and under stress he goes back to some old program like «How do you feel about that?» then you can interrupt and ask the same three questions: 1) «What step are you on?» 2) «What is the specific outcome you are going for?» 3) «How is what you just did going to achieve that outcome?»

If, in fact, that behavior wouldn't get the outcome, then as meta–person you ask «How could you get that outcome?» When he gives you one way, ask «How else could you get it?» When the programmer has three ways, have him go ahead and pick one to try out.

If the programmer is incongruent in delivering the method, you interrupt again. This time give specific feedback about what the programmer could do to be more congruent. «Change your voice tone and tempo in this way," or «Change your body posture and gestures in this way.» You are all here to become more graceful communicators than you already are. If there is any incongruity in your behavior, I believe that you'll want to know about it, because being incongruent is self–defeating. When you are the programmer, your 7 ± 2 chunks of conscious attention will be involved in communicating with the client and getting responses. The meta–person will have more attention free to notice what's going on, so use the information that the meta–person communicates to you.

As a meta–person the finest thing that you can do to assist the programmer is to interrupt at any time that you don't understand what's going on, the programmer hesitates, or the programmer is incongruent in his behavior.

Rose: So when my meta–person interrupts, rather than treating him like a mosquito and swatting him, I'm to treat my meta–person as a generator of new behaviors especially built in for me?

He will be a generator of new behavior only in the sense that he will challenge you by asking you the questions. You need to come up with your own solutions. He is not there to provide you with solutions directly.

Woman: Earlier when we were doing an exercise, the meta–person kept jumping in and interrupting what the programmer was doing. It seemed like the programmer was ready to go along to the next step, but the meta–person would jump in before the programmer could move on. Should I ask the meta–person to slow down?

Negotiate with your meta–person about the interventions that are appropriate for you. Keep in mind that it's difficult to interrupt too much in an artificial situation like this one that is set up solely for the sake of learning to have more choices. However, you are a human being with your own needs. If there are so many interruptions that it disorients you, say «Hey! I need at least a minute and a half at a time before you jump in again, unless there's something really important.» So negotiate that with your meta–person. And you can also reframe yourself so that you think of every interruption in terms of what you may be able to learn from it.

When you are the role–player, I want you to role–play your most difficult clients. Don't get into your own personal change goals. You'll get changes anyway, by metaphor. You don't have to worry about that.

I'll role–play this exercise again, to make it really clear. Doris, you be the programmer. What I'm going to do is think to myself «Oh yeah. I have one client who is really tough for me to reframe. Every time I do reframing, it starts out really nicely, but by the time I get to step three or four, the signals start shifting all over the place and I don't know what's happening. I don't know what to do with that.» So I tell Doris «You've got good rapport with me. You've helped me identify a behavior to reframe and you've established communication with the part. The signal I'm getting is an increase in heat in my hand for 'yes' and a decrease for 'no.' Now you're on step three and you're about to have me ask the part if it will let me know in consciousness what it's doing for me that's positive. It's there that the difficulties come up, so let's start there.»