The author's main disagreement with Harold Blum is over Blum's contention that symptomatic improvement is directly linked to the recovery of memories. The idea that memories are laid down in childhood and preserved until the time of their later recovery flies in the face of what we now understand as the creation of memories by the neurobiological systems underpinning this aspect of mental function. No evidence directly links symptomatic improvement to reconstruction and thus to outcome; care should be taken to avoid confusing co-occurrence with causality. While reconstruction of how things actually were in childhood can significantly contribute to therapeutic action, it is the process rather than the outcome of this reconstruction that is therapeutic, due to the opportunity thus afforded to rework current experiences in the context of other perspectives. The author clarifies his definition of transference to show some areas of agreement between his position and Blum's. He disusses contemporary neuroscientific views on memory and identifies a number of psychoanalytic writers who have used these productively