Developing at long lenght the case of a young bulimic patient around mutative periods of the transference relationship and of the fruitful drawbacks of a lateral transference within the cure, I try to stress the particular and specific characteristics of her objet relation. I suggest that one should interpret the addiction to food as being the problematics of flaws in archaïc identitification processes linked to the fact that the primary cathexis were “ to identify badly ” or were “ badly identified ”. The object could thus not be intrejected but only incorporated. Hence an endless quest (addiction) not so much of the object itself but, rather, an attempt to identify “ that very ” object in order to identify oneself to it and thus be able to avoid being alienated by it.