I'm sorry if this violates the rules of this list (I'm new), but the reason I am double posting this is because I feel that while the following definately concerns the psychic phenomena thread, it also opens up more general questions about codeing and epistemology. But, I am ready to stand and be corrected... I think the answer to your question, Dr. Grinder, is no.  It is not possible to find a psychic whos behavior can not be coded in non-psychic terms.  But I think the answer is no mainly because of the presuppositions that go into our understanding of behavior.  Fortunately, you (and others) in NLP vol I, make it very clear that NLP expands the general notion of behavior so that it includes what we do 'on the inside'.  However, and you may disagree with me here (or everywhere!), I believe that what people do 'on the inside' is an extremely unmapped portion of human behavior. NLP has provided what I consider to be the most effective tools for doing this sort of modeling, which have indeed already opened up a set of significant distinctions (sub-modalities for example)to carry out such a mapping task.  Yet, from the two genuine psychics I know, and from my attempts to get at what they are doing, I am certain that a new vocabulary is necessary to accuratly code what they are doing, and they sure as hell aren't able to provide it- just as milton, virginia and fritz weren't able to provide the necessary codeing for their magic.  I'm not writing this as a challenge to your statement, but I was struck by what I consider to be a sort of tautology within which it seems to function.  I must say that I definately agree with one implication of your question in that I think the terminology you find offered in the psychic lexicon is utterly inadequet for such a modelling task.  I'm studying the epistemological works of Rudolf Steiner and those who are operationalizing this epistemology via phenomenon centered science (or Goethean studies), and I'm finding some very valuable observation that are providing good clues as to the formal properties of such a desired codeing. Ben