Thank you for taking the time to write such a thoughful contribution to this thread and generally elevating the tone of the discussion. (I can only hope that I don't wind up dragging it down again.)

And you certainly win the sci.psychology vocabulary award for the month. ``Nomothetic'' and ``ideographic'' made me resort to the ten pound unabridged dictionary. I was trying to relate ``ideographic'' to the idea of Chinese ideographs, where often a concrete object stands for an abstract concept. But finally I realized that you really meant ``idiographic,'' from the same root as ``idiosyncratic.''

For the record: Idiographic -- specific, particular. Nomothetic -- general, universal.

> The variance in our
>research or clinical subjects is too great to allow any generalization
>from the case approach, and, conversely, irrespective of the detail and
>breadth of nomothetic studies, each client or animal carries its own
>variance with it, and should be seen as an individual.

My NLP teachers certainly agreed about the variance in subjects. In fact (and here I dip into Ross Perot mode for a moment) I remember Leslie Cameron Bandler saying to us ``When I ask students what they find most valuable in our seminars, there are lots of answers I get that are very satisfying to me. But there's one that I especially listen for and am always especially thrilled to hear: 'This seminar has helped me realize how incredibly different people are from each other.' ''

But NLP draws a different conclusion from this variance. First, since people are so incredibly different, statistically-based methods can have only limited value as a guide to a therapist. And secondly, since people are so very different, if one looks for only nomothetic information in one's research then one will often overlook the most valuable information one can get from a subject -- information about all the things that make that subject different from the rest of the world. And only by paying attention to such idiographic information from a number of subjects can one finally reach nomothetic conclusions of true depth.

>As I presented to a class once, it is the duty of the clinician to apply
>the scientific method. Yes, I said duty. The scientific method is simply
>the development of hypotheses about something, and then testing them.
>When a client enters the door, the client is a (series of) experiment(s).
>The root problem and best Tx is something that must be deduced by the
>clinician from observation and questioning of the client. This is the
>scientific method.

Are you sure you don't have an NLP background? :-) You sound just like Richard Bandler or some of the other leading NLP figures.

> This does require lots of education on the part of the
>therapist. If you are simply into NLP, dynamic psychology, RET, or
>whatever, you essentially are applying your particular hammer to every job
>that comes accross the door. Some hits, yes, but so what. You did not go
>into this field to stroke your ego, now did you? (I will keep my
>anecdotal answer to myself.)

This still sounds like what is taught in NLP seminars. It has to be admitted, though, that every therapist, whatever their education, is biased toward a particular set of tools. And lots of practitioners out there who ``do NLP'' have a very limited set of tools which they use on everything. Sometimes these work and sometimes something else would be better, maybe something that's not even NLP.

In particular, any therapist working with seriously mentally ill clients needs to know, in my opinion, a lot that's not covered at all in NLP trainings.

As far as I can tell, most conventionally trained therapists who go through NLP training (and there are many of them) do not simply throw away everything they used before. Who ever said that using NLP means that you're not allowed to use anything else?

>Finally, as I presented to a different seminar, we should all take a good
>read of Meehl's work, and others that followed. Essentially, the wonders
>of clinical judgement pale when tested against even a univariate
>regression equation. My inderstanding of this is that the refusal of
>clinicians to work to get this data down to the point of true usability
>essentially is the art part of the field. Clinical judgement, like art,
>feels good to do. Hence, clinicians do it. Hopefully it was done as if
>it was a statistical prediction equation. That way, more good can be done
>to more people, rather than just treating more people - some of whom fell
>outside the success window of other therapists.

The idea of NLP is to try to accomplish changes that are very specific and testable. If you are treating someone with a fear of heights, you can have him go up to the 20th floor afterwards and look out over the balcony. If you are helping someone to stop biting her nails, you don't need a sophisticated diagnostic instrument to test your success.

A big part of the art of doing NLP is to suitably ``chunk down'' a client's presenting complaint and create specific outcomes, stated in the positive.

A client seldom comes in and tell you what they want. They tell you what they don't want: ``I want to stop being angy at my husband/being indecisive/being such a wimp, etc.'' The NLP practitioner should respond by asking ``How would you like to feel/think/act instead?'' This is seldom an easy question to answer, and the skill of the therapist lies in helping the client to find that answer. Without that first step, the NLP techniques are often of limited value.

In the NLP trainings I've been through there was much more emphasis on finding specific positive outcomes than in the use of various techniques. There was an implication that in many cases simply finding such an outcome might be sufficient to solve a client's problem.

I feel compelled to offend you by putting a piece of idiographic information in here. In the videotape Lasting Feelings, the client (Hazel) begins by telling Leslie Cameron Bandler how jealous she is about her husband, even though she knows her jealousy is never justified, and how terrible she feels about that. In the work Leslie does, it's fairly clear that she has formulated the following positive outcome: ``I want Hazel to have ways of knowing that she is loved by her husband even when she sees him interacting with another woman. Furthermore, I want her to be able to discriminate between situations where jealousy is really warranted and those where it is not, and if she notices that indeed there is a real threat present, I want her to be able to deal with that threat resourcefully.''

In my experience, it is the conventionally educated therapists who often operate from sweeping generalizations such as ``Jealousy is a sign of insecurity'' or ``Jealousy comes from low self-esteem.'' This is being nomothetic at the price of being able to notice what's happening with the individual client.

I think that one of the reasons why there's never been developed (to my knowledge) any standard NLP approach to dealing with depression is that it's too hard to chunk the problem down and find specific positive outcomes.

I fully agree that no conclusion is possible for this discussion. For me, that is not really the question. For me, the issue is whether people can find value in a discussion whose purpose is not to reach conclusions but simply to explore ideas and phenomena.

This has always been what frustrates me when NLP has been discussed in sci.psychology. The academics seem to be interested only in coming to instant conclusions and are unwilling to hear about anything new unless it has already been thoroughly researched. To me, this attitude seems very unscientific.

Does finding value in the nomothetic mean that we have to totally reject the idiographic? Is it not possible that the two different approaches are actually useful for different purposes? That while statistical studies are useful for making assertions which are true in sweeping generality, that at the same time sometimes individual case studies can be more useful as a way of gaining insight or finding new directions to explore?

Isn't it possible that while the nomothetic approach is the ideal in science, the idiographic has almost universally been found the most effective in teaching an art?

--
When the main justification that a science has for itself is how scientific it is, rather than how many worthwhile discoveries it makes, that is a strong indication that something is wrong.