One book that NLP people seem to frequently refer to is Metaphors We Live By by George Lakoff and Mark Johnson.

> ....
>We used this type of speech as the building blocks for metaphor. For
>example a client might have said, ``It was an explosive situation.'' The
>figurative speech element is the word ``explosive.'' The client choose
>that particular word for a reason as opposed to other possible words (so,
>we theorized). By stringing together the elements of the figuative
>speech an unique speech pattern and metaphor ``landscape'' began to emerge.

NLP uses the term metaphor with a slightly different shade of meaning. I believe that a more correct term for what NLP calls metaphor would be ``parable.'' (It sometimes seems to me that the founders of NLP had a genius for using words in not quite the correct meaning.) The original NLP group learned about the power of metaphor (or parables) by observing Milton Erickson, who would often, instead of overtly discussing a client's problem, tell the client some long involved story apparently having no relevance at all. The original NLP book on the subject is Therapeutic Metaphors by David Gordon. Most introductory books discuss the technique at least briefly, for instance Introducing Neuro-linguistic programming by Joseph O'Connor and John Seymour or Practical Magic by Steve Lankton.

>This is where metaphor gets potentially dangerous. Metaphors are
>potentially powerful tools. It is suspected by many in the field that
>because metaphor is ``one-step'' away from the object actually being
>discussed the client's defenses are lower.

By-passing client resistance is usually the reason given in NLP for using metaphor. I've never heard NLP trainers warn against metaphor as a dangerous technique, but it certainly makes sense that there could be dangers.

> The clients continued to mull and
>mutate the session's metaphor after leaving the session. Depending upon
>the direction that the therapist left the metaphor the client mutated the
>metaphor in good and bad directions. If it mutated in bad direction, we
>had the potential for crisis and afterhours repair work on our hands.

To repeat, NLPers apparently do not experience the phenomenon of having a metaphor mutate in a bad direction. Perhaps this is because NLP metaphors are constructed a little more carefully than described above. Or it's possible that NLPers are simply not sufficiently aware of the dangers.

In NLP, it is usually taught that metaphors address the unconscious mind rather than the conscious mind, and in fact the client is usually not intended to consciously figure them out. (Rather than ``mulling them over,'' the intention is that the client not even consciously think of them at all.) As indicated above, it usually takes few weeks for the effect of the metaphor to become manifest, which makes metaphor different from most NLP techniques. (This also makes it more difficult to be certain when ascribing a change in the client to the effect of the therapist's metaphor, since other things will have been happening in the client's life in the intermediate time. Of all the NLP approaches, I think that metaphor is the one where statistical evaluation of effectiveness would be most useful. It is also, however, the technique which is most difficult to evaluate statistically, since so much depends on the artistry of the therapist.)

Metaphor is one of the NLP techniques I feel least comfortable about, partly because I've just never been much good at constructing therapeutic metaphors.

To conclude, I will give an example of a very brief metaphor, constructed by one of my trainers for a student who had difficulty feeling comfortable in groups.

``Once upon a time, it used to be true that green just didn't seem to fit in with the other colors. All the other colors seemed to harmonize together nicely and make a pleasant pattern, but green was always the oddball. One day green was sitting in a corner, away from all the other colors, feeling left out and unhappy, and someone came over and said, 'You know, if you'd just slip in between yellow and blue, you'd fit right in.'

``And so green did that. And ever since then .... there've been rainbows!''

--
Unlike past American intellectuals, who saw the educated nonacademic public as their main audience, today's leftist intellectuals feel no need to write for a larger audience; colleagues, departments, and conferences have come to constitute their world.     -- Russell Jacoby