As I have already demonstrated, scientific discourse about anorexia tends to ascribe a very specific place and function to sociological analysis. The latter is generally tasked with accounting for the ‘recruitment’ of anorexics in historical terms – why today? – and in social terms – particularly related to gender, social class and age. In other words, its role is to deal with the epidemiological dimensions and social context of the illness. The most obvious way of going about this ‘sociologizing’ consists in taking quantitative epidemiological data and then using interviews to try to explain the specific boundaries of this recruitment. This is the task that social historian Joan Jacobs Brumberg assigns to the social sciences, drawing a distinction between analysing recruitment – which requires sociological and historical approaches (what historical and social conditions mean that the disease appears at a given moment in time and in a specific part of the social space?) – and analysing individual careers, which should be left to biology and psychology:

An individual may begin to restrict her food because of aesthetic and social reasons related to gender, class, age, and sense of style. This constitutes the initial ‘recruitment’ stage . . . . An individual’s dieting moves across the spectrum from the normal to the obsessional because of other factors, namely emotional and personality issues, and personal physiology and body chemistry . . . . Anorexia nervosa can be conceptually divided into two stages. The first involves sociocultural context, or ‘recruitment’ to fasting behavior; the second incorporates the subsequent ‘career’ as an anorexic and includes physiological and psychological changes that condition the individual to exist in a starvation state. The second stage is obviously the concern of medicine and mental health professionals because it is relatively formulaic and historically invariant. Stage one involves the historian, whose task is to trace the forces and events that have led young women to this relatively stereotypical behavior pattern.