Deconstructing behavioural classifications: tobacco control, ‘professional vision’ and the tobacco user as a site of governmental intervention

Abstract

In this article, I examine a defining feature of the ‘new public health’: the (re)construction of health-related phenomena in behavioural terms. While the ‘behavioural turn’ within epidemiology has had far-reaching implications for the way in which public health problems as a whole are conceptualised, including, significantly, obesity and alcohol (mis)use, here I explore how the new public health works up its behavioural objects using the example of tobacco use. Beginning with the work of counting smokers, I trace the emergence and consolidation of a standard model for identifying and measuring tobacco-related harm, a model, I argue, that has been extended so that tobacco use itself can be treated in disease terms. As I show with reference to an example of contemporary public health research practice in the UK, this extension is problematic because it establishes a depoliticised view of the public's health that concentrates on individuals, recast as bundles of problem behaviours, at the expense of any examination of the social, cultural and economic circumstances in which those individuals live. Epidemiological research of this kind, with its core message that behavioural problems require behavioural solutions, relies on close alliances between the health sector and decision-makers more broadly. Under these conditions, the point at which research ends and government begins is often difficult to locate. I conclude by arguing that we should pay greater attention to the epidemiological practices used to transform the behaviour of the tobacco user, like that of the eater or drinker, into a site of governmental intervention.

Keywords:

• models of the tobacco user
• the new public health
• health and government

Notes

Notes

1. As one reviewer pointed out, it can be difficult to detect the effects of this subtle shift because it has operated largely at an infrastructural level. Perhaps its most visible index, however, is the near ubiquitous recourse to behavioural classifications within contemporary public health, with drinking, eating and smoking, for example, routinely transformed into ‘drinking behaviours’, ‘eating behaviours’ and ‘tobacco use behaviours’ for the purposes of analysis and intervention. These translations from ordinary to (pseudo)technical terminologies are not merely so many loose ways of talking but constitutive features of much contemporary public health practice (Mair and Kierans (2007) for an examination of how this has played out in the field of tobacco control).

2. The question of what ‘counts’ as the first scientific evidence of the link between tobacco use and lung cancer has been the subject of recent controversy. As has now been established, clinical epidemiologists from the 1920s on, many working for the Weimar and Nazi regimes in Germany, had reached similar, though less rigorously evidenced, conclusions (for commentaries see Smith et al. 1994, Proctor 1997, Smith 2005; for a ‘corrected’ timeline of discovery see Proctor 2001).

3. The participant's name has been altered.