GSM, I read an article a few years ago which told of a number of empirical studies which proved that Freud's notions about dreamwork had a large degree of credibility.  But this was only after half a century or more of other empirical studies proving the opposite, that Freud's theories had no basis in reality.  The earlier research led to the popularization of both behavioral and cognitive schools, and the demise of psychoanalysis.  As close as NLP often seems to be to the clinical psych / cog sci fashions of our times, at times, it is really closer to psychoanalysis in that it is based on intuitions and observations.  While the scientific method can get you to the moon faster than superstition, I have my doubts about whether current scientific methods will ever show themselves more effective than the alternative approach.  Let Erickson be the prime example.  Remember, you're dealing with consciousness, which can never get outside of itself.  In the end, objective science applied to behavior is as superstitious as anything else, and a great degree of enlightenment wishful thinking. I think that in most of the best of NLP and Ericksonian texts the authors make mention that there is no such thing as the "unconscious," that it is an abstract model that is useful.  Cog Sci would not even approach it, since there is no way to accumulate data based on such a fiction.  But that assumes that an epistemology of piecing things together (small chunk) is more reliable than constructing myths based on intuitive visions and observation (large chunk).  One is the accountant's approach, and one the madman's.  But in the end, who's madder? best, JS