Manipulation and Population Statistics in Nineteenth-Century France and England

Social Research, Summer, 2001 by Libby Schweber

An examination of the history of the Statistical Society of London (SSL) and the public health movement suggests that in the next few years Neison won the debate. Within the SSL Neison was one of a core of younger members, including William Guy and William Farr, who took over the intellectual and organizational leadership of the society during the first half of the 1840s. An examination of articles published over the next 10 years reveals the widespread adoption of Neison's model as the authoritative form of social statistics. While most authors lacked Neison's mathematical skills, their articles are peppered with tables they had asked Neison to construct for them.

Academy of Sciences, France, 1838

On the other side of the channel, mortality tables and population statistics more generally were the subject of two major debates. The first was played out at the Academy of Sciences in 1837 and the second took place at the Academy of Medicine 30 years later in 1867-1868. Like the 1844 debate in England, both of these exchanges involved a confrontation between different epistemologies and both combined technical, epistemological, and political considerations. In contrast to the discussion in England, no clear alignment of considerations appears in either discussion.

The 1837 debate involved an exchange of letters that were sent to the Academy of Sciences and published in their Transactions. The occasion was the competition for the prestigious Prix Montyon, a prize that the academy granted each year to the best work in statistics.(7) The occasion was the submission of a set of mortality tables by Mr. Demonferrand, a mathematics professor (at a lycee) and corresponding member of the academy. The work was entitled "An Essay on the Laws of Mortality and of the Population in France." As the title of his essay indicated, Demonferrand, like many of his contemporaries, associated mortality tables with a particular use of statistics aimed at the identification of otherwise invisible regularities or "laws." The exchange was provoked by a letter from Alexandre Moreau de Jonnes, the director of the newly created national statistical bureau, the Statistique Generale de la France (SGF), to the academy asking it to refrain from awarding Demonferrand the prize.(8) As in England in this period, statisticians were ambivalent about the scientific status of mortality tables. And as in England, their differences point to two alternate epistemologies that were under negotiation.

To appreciate the rivalry between the two men it is helpful to know that Demonferrand's study had originated with a request by the Academy of Sciences to evaluate the national census of 1835 that Moreau deJonnes had produced. It is also important to keep in mind that the scientific status of national-level data was far from established. The Academy of Sciences defined statistics as certain knowledge grounded in a personal knowledge of local conditions. It was this familiarity with the local population and direct involvement in data collection that assured the reliability of the original data and rendered statistics scientific. National official statistics failed to meet this criteria; studies using official data were consistently criticized for using second-hand data by the committee for the Prix Montyon.