L’étude des écrits médicaux anciens offre, au même titre que les études transculturelles, la possibilité aux psychiatres de reconnaître aux troubles mentaux des expressions constantes ou au contraire variables qui orientent vers une pathogenèse de nature plutôt biologique ou plutôt socio-culturelle. Cet article examine la question délicate du diagnostic rétrospectif et de la catégorisation pathologique des troubles mentaux dans les écrits médicaux anciens, et tente d’y répondre d’une manière pragmatique et opérationnelle. Une méthodologie spécifique, psychiatrique et historique, a été mise au point et appliquée à un corpus de plus de 2000 consultations écrites françaises des xvi^e siècle et xviii^e siècle. Les résultats de cette application sont illustrés par l’analyse détaillée d’une dizaine de cas et discutés dans leurs implications pour la recherche psychiatrique historique.

Like cross-cultural studies, research on ancient medical writings allows psychiatrists to identify constant and variable expressions of mental disorders; this may in turn allow discrimination between pathogenesis mainly of a biological nature and that mainly of a socio-cultural nature. However, the study of these writings presents many problems which require the development of specific and rigorous research methodologies. In particular, close attention is needed for the identification and characterization of mental disorders, a process which is usually referred to as “retrospective diagnosis”. This paper tackles the difficult problem of retrospective diagnosis and pathological categorization of mental disorders described in historical medical writings and proposes a pragmatic and operational approach to these issues.

We propose going beyond the simplistic contraposition of essentialist and constructivist approaches to mental disorders. First, history research questions involving individual diagnosis, either to ascertain whether a given disease or nosological entity is present in a given historical population (situations A1) or to understand a behavior of an individual or group in a given historical context (situations A2), should be distinguished from questions in which the health of a population or a subgroup is of interest (situations B). Situations of type (A1) require the use of all medical knowledge to make a tentative retrospective diagnosis; situations of type (A2) require considering retrolective diagnoses which could have been made at the time the disorder was managed/reported; and situations of type (B) require using robust pathological categorizations and classifications like those used in contemporary epidemiology to allow comparisons between populations. This conceptual and methodological framework was applied to the study a corpus of more than 2000 reports of consultations of French physicians, written during 16th–18th centuries. Retrospective diagnosis was attempted by a group of psychiatrists and historians on the basis of DSM-IV-TR diagnostic criteria for common mental disorders.

Mental disorders accounted for nearly 4% of the medical conditions dealt with in the consultations; there was no trend for change during the study period. Minor mental disorders, such as depressive, anxiety and somatoform disorders were the most frequent. Serious conditions, characterized by symptoms corresponding to the contemporary categories of bipolar disorder, major depressive disorder and schizophrenia, were also suggested.

This study evidences the presence, in the early modern French population, of many mental disorders fitting into the categories of contemporary psychiatric nosology. It also demonstrates the feasibility of pathological categorization, and in some cases retrospective diagnosis, of mental disorders described in historical medical writings. These findings have implications for historical research in psychiatry.