I once did some work with a lawyer who is a trustworthy person, but nobody trusts him. His non-verbal analogues are terrible; they make everyone suspicious. His problem was that he couldn't get clients to confide in him so that he could represent them well. And half the time he was court-appointed, which made it even worse. What he really needed was a complete overhaul in his analogue system. Rather than do that, I taught him a little ritual. He sits down with his client and says "Look, if I'm going to be your lawyer, it's essential that you trust me. And so the question that's really important is how do you decide if you trust somebody?" He asks "Have you ever really trusted anybody in your life?" and he sets up an anchor when the client accesses that feeling of trust. Then he asks "How did you make that decision?" Then all he has to do is to listen to a general description of their strategy: "Well, I saw this, and I heard him say this, and I felt this." Then he presents information back in that format: "Well, as I sit here, I want you to see blah blah blah, and then I say to yourself blahdeblah blah, and I don't know if you can feel this," and fires off the anchor that he made when the person had the trusting feelings. I taught him that ritual and it was good enough.

But there is a real difference between that outcome and the outcome that you're working toward as a therapist. Therapy is a much more technical business in the sense of changing things. As a therapist you don't need to be nearly as flexible in terms of utilization as somebody who's a lawyer. A lawyer must be a master of the art of utilization. You need to be able to do many different things in terms of eliciting responses. You have to get twelve people to respond the same way. Think about that. Imagine that you had twelve clients, and you had to get them all to agree when you weren't in the room! That's going to take skill.

One thing you can do is to identify the one or two individuals, or several, on the jury who might, by virtue of their own strategies, persuade the others to go along. And of course that is what family therapy is all about. Everything is going to interact in a system. I don't care who you put together for what length of time, the systems are going to start clicking. I try to figure out who in the family elicits responses the most often. Because if I can get that one person to do my work for me, it will be really easy. Very often it's someone who doesn't speak much. Son here says something. He has external behavior. And when he does, you get an intense internal response from the mother. Although her external behavior is subtle, some little cue, everybody responds to it. When the father does something with external behavior, this kid responds, but not much else happens. And if the daughter does something, maybe we get a response here and maybe there.

I want to know who everybody else in the family responds to a lot. I also want to know if any one single person in that family can always get that person to respond. Let's say every single time the son does anything with external behavior, the mother responds. If I can predict something about how that happens, I can make one little change in the son, and then the mother will respond and get everybody else in the family to respond for me. I always spend fifty percent or more of whatever time is allotted to me gathering information, and testing it to make sure that I'm right. I’ll feed in an innocuous thing here, and predict what will happen over there. I keep running the system over and over and over again until I'm absolutely sure that if I make a change with this kid, it's going to change the mother's behavior in a way that will change all the other people in the family. That will set up a new stable system. Otherwise you usually get an unbalanced system, or they change in the office but they go home and go back to normal. I want something that's really going to carry over and be very, very permanent.

If I can set up a stable system by making only one change, it will be very pervasive with a family system. I think the main mistake of all family therapists is that they do too much in a session. If you're working with an individual, you can do a thousand things and get away with it, unless they go home to a family. One of the first things I always ask people when they come in is "What is your living situation?" because I want to know how many anchors I have to deal with at home. If they live with one other person, it's not so bad. You've just got to be careful that there's no secondary gain: that they don't get rewarded for whatever behavior it is they want to change.

Man: How much dependency on you is created by your methods?

One of the things we strive for in our work is to make sure that we use transference and countertransference powerfully to get rapport, and then to make sure that we don't use it after that. We don't need it after that. And since they don't get to sit there and tell us their problems, we don't become their best companion. There are real risks in doing content therapy because you may become someone's closest friend. Then they end up paying money to hang out with you because no one else is willing to sit around and listen to them drivel about unpleasant things in their life. We don't get much dependency. For one thing, we have a tool that we teach our clients to use with themselves, called reframing, which we are going to teach you tomorrow.

If you ask the people who were up here for demonstration purposes, my guess is they would assign very little responsibility to us for the changes that occurred in them—much less than they would in traditional content-oriented therapy. That's one of the advantages of secret therapy. It doesn't create that kind of dependency relationship.

At the same time, people who work with us usually have a sense of trust; they know that we know what we are doing. Or they may be totally infuriated with us, but they are still getting the changes they want. And of course we work very quickly, and that reduces the possibility of dependency.

In our actual private practice, which is severely reduced now because we're moving into other areas of modeling, we tell stories. A person will come in and I don't want them to tell me anything. I just tell them stories. The use of metaphor is a whole set of advanced patterns which is associated with what we've done so far. You can learn about those in David Gordon's excellent book, Therapeutic Metaphors. I prefer metaphor artistically. I don't have to listen to client's woes, and I get to tell very entertaining stories. Clients are usually bewildered or infuriated by paying me money to listen to stories. But the changes they want occur anyway—no thanks to me, of course, which is fine. That's another way to make sure there is no dependency. You do things so covertly that they don't have the faintest idea what you are doing, and the changes they want occur anyway.