Labelled Bodies: Classification of Diseases and the Medical Way of Knowing

Abstract

WAYS OF KNOWING IN NORMAL STIVIIn the 1970s, with the practical turn in the history of science, many historians influenced by science studies abandoned the concept of a unified 'science' governed by a single 'scientific method' and replaced it with the plural 'sciences'. The focus shifted to investigations of the distinctive research practices of each scientific field.1 Historians of science also broadened the scope of their investigations and became interested in practices in the laboratory and the field; material, social and discursive technologies; organization of scientific labour; circulation of instruments, reagents and documents; the roles of women in science; or science in non-Western countries.2 But in spite of these post-Kuhn, post-S SK changes and the integration of approaches from anthropology and cultural history, the new 'science studies' often continued to focus on university and academic setting, and to put 'pure' science above 'applied' . Scholars still concentrated on cutting-edge science, discoveries and major changes, and tended to neglect more mundane issues and practices. As John Pickstone argued in 2007, the Ways of Knowing (WoK) perspective which he developed to span science, technology and medicine may help us study more mundane ways in which scientific knowledge is produced and diffused, including those in routine clinical and industrial settings.3 My text looks at one of these mundane sites of production of knowledge: the clinic/hospital. It examines the classification of complaints and symptoms, including the use of paper technologies - and more recently electronic tools - to produce data, differentiate 'normal' from 'pathological' , manage sick and healthy bodies, and organize routine medical labour.Classification has sometimes been presented as an autonomous way of knowing. Ordering of variety by comparison and taxonomy is one among the six styles of scientific reasoning proposed by Ian Hacking, following Alistair Crombie. It is closely linked to another style of scientific reasoning, statistical analysis of populations, because in many cases analysis relies on the recognition of distinct categories.4 In the WoK perspective, classifications are an important part of 'natural history', a term used in its widest (and Baconian) sense, to include the description, collection, classification and display of any kind of object or process.5 Various kinds of analysis can feed back into taxonomies, different kinds of analysis can give different taxonomic outputs, and these may need to be reconciled with each other and with the earlier, pre-analytical natural history. And scientists and practitioners continue to elaborate new working taxonomies - a new 'natural histories' which we might call post-analytical.6 This stress on the interactions of numerous ways of knowing is well exemplified in medical work, where complexity and 'impurity' are the rule, and intervention (therapeutics) is always intimately entangled with representation (diagnosis and classification).This article explores the role of classification in clinical work. At its centre is the claim that nineteenth-century medicine saw the coming together of three distinct developments: the elaboration of the concept of specificity of diseases, the wide diffusion of new paper technologies and administrative practices, and the use of statistics to define normal values for a population. Together, they made up a new and powerful entity which we can call the 'disease classification system' (DCS). The DCS may be seen as new form of natural history (or specifically of nosography), fed by new forms of analysis (anatomical, physiological and statistical), and using new forms of data management. In the early nineteenth century, clinical knowledge produced by the Paris Clinical School and its followers was closely related to the analytical knowledge produced by chemists, and those generated within museum collections. …