Attempts at satisfactory psychoanalytic theory are as old as our field. Analysts have developed vocabularies that refer to unobservable entities, at least, to catalog their findings and, at most, to predict events through arguments about these entities. Although it has long been clear that the major contributions of psychoanalysis to the study of man and therapeutics are its clinical findings, their categorization and generalization, analysts have worked hard to coax the "witch" metapsychology into the light of abstract theory.

The history of virtually all sciences shows that the connection of observable phenomena and theoretical formulations is complex. Especially in disciplines such as psychoanalysis, with massive amounts of data, even communicating findings with reasonable brevity requires a technical, theory-grounded, vocabulary.

By the late 1970's it became clear that metapsychology not only failed as an abstract psychological theory, but that its language actually