Responds to objections raised by C. F. Halverson concerning the retrospective method used by the present authors in their study of recalled parent-child relations and personality. It is argued that retrospective accounts can provide a useful source of evidence when the probable effects of biases are considered.