Magic and Schizophrenia: By Géza Róheim. Posthumously edited by Warner Muensterberger with assistance of S. H. Posinsky. New York: International Universities Press, Inc., 1955. 230 pp.

In this posthumous publication Róheim compares the origins of magic and schizophrenic fantasies—a topic with which he was preoccupied in many of his writings. The similarities between magic and compulsive rituals have been pointed out by Freud, and dealt with in greater detail by Abraham in his early works. The basis of both lies in the omnipotence of thought which the child as well as primitive man possesses, while the repressed ego of the neurotic and psychotic substitutes this omnipotence of thought for realistic mastery of the environment upon which the gratification of instinctual needs depends. This concept is by no means new; the contribution of Róheim consists primarily in its wide documentation with anthropological observations and a fine study of the fantasies and dreams of a schizophrenic patient. Róheim, however, goes farther than merely pointing out the similar nature of these phenomena. His thesis is that the first object relation of the child is to the mother; hence, 'object relations in general are based on relations to the mother … mother is the prototype of Mother Nature'. Since the original object relation to the mother is oral, object loss means the loss of the feeding mother. Magic and schizophrenia are attempts to re-establish the lost mother-child unity by introjection and identification.

Róheim reaffirms Rado's earlier formulation concerning the function of oral gratification in ego development. According to Róheim, the oral function 'is one of the earliest and most important nuclei of ego development'. He then tries to account for the differences between magic and schizophrenia. Magic proceeds in this fashion: '1, originally there is an object-directed cathexis; 2, then withdrawal and a secondary narcissism; 3, then a movement back from the "autarchic" position, a return to object. These processes in magic are egosyntonic and shared by the social group. In psychosis these processes are not egosyntonic and are not shared by the so