Manipulation and Population Statistics in Nineteenth-Century France and England

Social Research, Summer, 2001 by Libby Schweber

The differences between Chadwick and Neison centered on the choice between different measures of what Chadwick referred to as "actual" mortality and its causes. Both authors objected to the usual method of measuring mortality by the general (or crude) death rate (deaths over total population). Chadwick argued for the substitution of the average age of death for the general rate, while Neison called for the construction of special mortality tables, by district and by social class. For Chadwick, the advantage of the average age of death was that it incorporated the effect of population growth, providing a higher and thus more realistic picture of the ill health of the population. For Neison, both measures were flawed in that they failed to take into account the effect of the age structure on the measure of mortality. The advantage of special mortality tables was that they corrected for these distortions, thereby providing a better basis for comparison across cities.

To illustrate his point, Neison performed numerous statistical experiments aimed at demonstrating the distorting effect of age structure. Taking two districts with different average ages of death and with different age structures, he calculated the average age of death of one if it had the age structure of the other and, conversely, if it had its own age-specific mortality rate but the age structure of the other. The point of the demonstration was to show that "practical applications produce contradictory results" (Neison, 1844: 45).

Chadwick, however, dismissed such concerns as involving a confusion between cause and effect. According to Chadwick, the proportion of children or old people in the population is a result of the mortality rate. He thus objected to treating age structure as an independent factor. As for mortality tables, he rejected them as general abstractions divorced from the actual experience of a specific population. Instead, he argued "that the actual experience of any class, or place or period ... is a safer guide than any insurance table deduced from the experience of another people living at another time and place, or any assumed general law" (Chadwick, 1844: 8).

The differences between Neison's and Chadwick's positions can, in part, be related to differences in their concept of population and of mortality as objects of inquiry. For Chadwick, these were empirical constructs, constituted by individuals. This position is evident in Chadwick's continual use of the term "actual" to describe mortality rates as well as causes and in his principled criticism of "hypothetical reasoning." As he insisted, "it is only by examinations, case by case, and on the spot, that the facts from which sound principles are deduced may be correctly distinguished" (1844: 29). For Neison, mortality was a force or law that could be identified only if it was abstracted from the variety of other elements at play in the production of the observable population.

In the course of the debate both authors explicitly linked their technical and epistemological positions to the uses they envisioned for their preferred measure. For Chadwick, the purpose of the average age of death was to establish the relative salubrity of local districts and, more important, to establish the role of physical conditions--problems of drainage, sewage, ventilation, filth, sanitation--in the local health conditions. As he wrote, "the actual morality of most districts is found to be coincident chiefly with its physical condition, and is most accurately measured by the years of vitality which have been enjoyed, i.e., by the average age of death" (1844: 6).