This article explores Alfred Lorenzer's notion of scenic understanding through a conflict that beset a data-interpretation panel, part of a study evaluating a programme of psychodynamic therapy for drug-addicted young people. A second interpretation panel, formed in the wake of the first group's collapse, came to understand this conflict as a ‘counter-transference’ to the latent meaning present in the case on which the first panel had been working. This insight, in turn, gave rise to a new understanding of the case itself, in particular the possibility that the interviewee had made less progress in her therapy than appeared to be so.