This study of Canadian nineteenth-century obstetrical and gynaecological case histories enters the current debate about narrative prose and its postulated relation to humanitarianism from a particular point of view. It uses literary analysis to generate new types of evidence to shed light on the development of this medical specialty, and so demonstrates how close textual analysis can be of use to medical historians. The finding that narrative structure played a central role in the development of statistical thinking about disease grants it a highly ambivalent role in relation to humanitarianism. It supports the conclusion that the literary form has no content of its own; narrative appears as a powerful vehicle for creating consensus, but this insight cannot commit one to a view about the subject of the consensus or its social implications.