Tuttman surveys the history of the "borderline" concept in A. Stern, H. Deutsch, Knight, Stone, Frosch, Kernberg, Mahler, Masterson, Rinsley, Abend et al., and Kohut. He feels that their findings all support the point that treatment requires that the analyst be able to accept and contain intensely loving and hating projections. He divides analytic approaches into two categories: a neutral, interpretive position, exemplified by Kernberg, and a nurturing, reconstructive experience, such as Kohut's. The former is necessary, but by itself it can be experienced by the patient as an accusation, and lead to hostile stalemates, which the latter is needed to undo. Tuttman believes that an object relations perspective provides an emphasis on both the intrapsychic, structure-forming aspect of objects and the world of human subjectivity. The analytic process involves both a real interpersonal relationship and a transferential, fantasy relationship based on past experiences.