We Must Publish Material from Our Sessions, So how do We do It Safely?

This paper discusses whether the Internet and related technologies that we can imagine in future may have made it unsafe for psychoanalysts to publish clinical material in an anonymised form. The author notes, first, that algorithms work with identifiers of various sorts to tie information subjects together across space and time. It means that what is an anonymised clinical report today might, in future, potentially be joined with other documents to allow a patient to be re-identified. Discussing the current literature from computer science, the case is made that while there will be risks, if anonymisation is done less than rigorously, they can be removed by a variety of measures, most important of which are both the generalisation and falsification of potential identifiers. Exploring the merits and ethics of this approach, the author argues that psychoanalytic data are not externally valid data but internally consistent subjective data. It means that ensuring external identifiers are absent or distorted in clinical reports to convey the analyst's understanding but to safeguard against jigsaw puzzle re-identification does not de-authenticate them. Indeed, the process of anonymisation, focusing on the status of psychoanalytic data as inherently subjective, should deepen and strengthen rather than reduce their evidential values.