Manipulation and Population Statistics in Nineteenth-Century France and England

Social Research, Summer, 2001 by Libby Schweber

In summary, two features of the debate appear important. The first concerns the very weak articulation of the different considerations at play in the debate and the exclusion of epistemological issues from the resolution of the debate. The second concerns the missed occasion for the introduction of a new style of specifically statistical reasoning and the consequent reinforcement of a particular descriptive model. Both features were directly related to the monopoly of the Academy of Sciences over the distribution of scientific recognition for statistics and the particular model for statistics the academy rewarded through the mechanism of the Prix Montyon. An examination of the reports of the academy over the course of the century point to the persistence of this sharp separation of mathematics and statistics and this definition of statistics as the craft of data collection.

Academy of Medicine, France, 1867

The fourth debate with which this paper is concerned took place in the context of a prolonged discussion at the Academy of Medicine in 1867. As in the other cases, the discussion combined technical, epistemological, and political considerations. But in contrast to the other debates, the discussion was not centered on the choice of a statistical measure, nor was it conducted in an organization explicitly devoted to statistics. Instead, the technical debate was part of a broader discussion on the state, or rather dynamics, of the French population.

In 1867 the minister of public instruction asked the Academy of Medicine to consider the problem of infant mortality and, more specifically, wet-nursing (nourrissons). The problem of nourrissons was one of a number of social problems that had been documented, and thus revealed, by statistics and, as one would expect, numbers figured centrally in the discussion. None of the academicians questioned the existence of the phenomena or its gravity; instead, the debate focused on the role of the academy. Some members believed that they should call for further scientific inquiry while others felt that the dimensions of the problem were already known and that the academy should provide the government with concrete suggestions for action.

Toward the end of the discussion, Dr. Jules Guerin, one of the leading figures in the public-hygiene movement, cited an increase in the number of military conscripts unfit for duty as evidence of the degeneracy of the French population. Dr. Felix Hippolyte Larrey challenged Guerin's argument and appealed to Paul Broca, a new member of the academy and founder of the Society of Anthropology, to establish the facticity of the claim. Broca accepted the appeal, but suggested reserving the discussion for a separate session. The first discussion on infant mortality was thus followed by an equally lengthy second discussion on depopulation.

Within the second discussion two speakers devoted a large part of their remarks to statistics and to statistical measures. These included Broca, whose opening presentation covered 53 pages, most of which involved a systematic consideration of different statistical measures used to evaluate depopulation, and Dr. Guerin, who, in an equally long response, countered most of Broca's arguments. These two lengthy presentations were followed by a series of shorter exchanges between the two men in which they criticized particular points in their opponent's exposition. By concentrating on the exchange between these two men and on the points directly related to the problem of how to measure population growth, we can glean insights into the state of statistical reasoning among public hygienists in France at the time and make a comparison with the other statistical debates examined earlier.