Forty-four years ago, as a psychoanalytic candidate, I began my fi rst supervised analytic case. It was before Kernberg (1975) redefi ned borderline conditions and suggested that they could be understood and treated dynamically. Using our current diagnostic criteria, my patient would be seen as having a borderline personality disorder. He was a 27-year-old man with symptoms of depression and anxiety, socially inhibited, and very bright. Th ree months after he began his analysis, as he increasingly experienced aff ects of safety with me because of my ordinary analytic attitude of kindness, consistency, and interest, a dramatic change developed that was unfamiliar and unlike anything that I had experienced in a dynamic psychotherapy. He began ranting and railing at me and telling me about some of his history for the fi rst time. He described masturbatory fantasies where he was having sex with his mother while stabbing his father. He would hit his chest and cut his leg while masturbating. His ranting contained the same content as the criticism that he frequently directed at his parents: I was inept, uncaring, and neglectful. At that time, I did not have a clinical model I could use or adequate clinical support from my supervisor to understand his actions or to respond analytically to him. I had only my intuition to help me make analytic sense out of my experience with him.