Disciplining Statistics: Demography and Vital Statistics in France and England, 1830-1885

As its title implies, Disciplining Statistics concerns itself with understanding how demography and vital statistics became institutionalized as disciplines in France and England. In this book, Libby Schweber focuses less on "competing forms of facticity" and more on differences in the process of institutional recognition and disciplinary consolidation. For her, then, it is not just the degree of fit between demography and the political agendas of the day that matters, but also the thoughts and actions of key proponents that need to be taken into account for understanding the formation of a discipline. Schweber aims to explain not only levels of success or failure, but also the timing and qualities of the discipline that ultimately emerges. For anyone interested in these issues, I recommend Disciplining Statistics.

Schweber agrees with her predecessors (Kuklick and Fuller, among others) that new disciplines, once formed, do not merely reflect a neatly carved-out space, a clear consensus about the importance of a new area, how that area should look, or whether the proposed discipline should even be taught within the academy. Instead, as a practice that links "technical, rhetorical, conceptual and social elements," it is the combination of factors that explains disciplines and, consequently, the differences in disciplines across countries.

In France, the relative omnipotence of the Grandes Écoles determined the validity of most things either academic or intellectual; this, when coupled with the line that was customarily drawn between scientists and public policy-makers, made it very difficult for demography to rise as a discipline. To gain legitimacy, demography's advocates had to sidestep academic institutions by establishing scientific journals and congresses, a task that Schweber shows to be challenging but eventually successful. In England, by contrast, debates were less centralized and philosophical, and they tended to focus on the political utility of vital statistics. Consequently, vital statistics were politically motivated and valued, and, even if they were not always cherished within the ivory tower, it was easier for them to survive in the face of scientific criticism.

In making this argument, Schweber aligns herself with Hacking and Porter, two pioneers in showing how national boundaries shaped early statistical thinking. She goes beyond her predecessors, however, by emphasizing the discontinuities. Demography, we are told, had a few false starts and periodically fell out of favour throughout the nineteenth century. Not until the late 1870s did it begin to receive institutional recognition in France. Vital statistics in England had an easier time, given their immediate recognition within policy circles.

The book does a good job presenting rich historical detail with an eye to keeping readers interested in what can be a dry topic. It is fairly well written and logically presented and organized, and technical jargon is used only when necessary. Given the topic, this was no doubt a difficult task. Despite the book's many strengths, a few questions for me remain unanswered (or at least under-answered) in Disciplining Statistics. First, why look only at England and France? What was happening in other countries at this time? If we accept that national censuses had at least something to do with demography (and vice-versa), which Schweber never really addresses, what was happening in countries like Denmark, Norway, and the United States, where some of the first censuses were conducted? Were there similar debates about the utility of the analytical methods and the consequent forms of knowledge occurring in these countries? If so, why do France and England form the basis for identifying how statistics were "disciplined"? Are these countries exceptional or illustrative of what was happening in Europe at the time? What about other parts of the world? Further, where were early demographers being trained in the absence of formal institutional recognition? Might some of these links be relevant for understanding how statistics were disciplined?

Secondly, what does identifying the process of institutional recognition provide for understanding disciplinarity that looking at competing forms of facticity does not? Some of this information appears in the introduction to the book, but is not carried through the text. Why should readers interested in discipline formation accept the framework in this book over earlier attempts?

Thirdly, although I see great merit in heeding Latour's invitation to "follow the scientist," I was at times left wondering why we were following the scientists in question. Why, for example, was Malthus not more central in understanding how demography came to be? Even though he taught political economy (rather than demography) while training civil servants at Haileybury College, a good deal of what he wrote (particularly Essays on the Principles of Population) might be considered early demography. Malthus is widely recognized as one of the western world's earliest demographers (try to find an introductory demography course that does not at least cursorily discuss Malthus), so why is he never mentioned (even in the bibliography)? Bruce Curtis, in his review of this book (Canadian Journal of Sociology Online, March-April 2007), has remarked on Schweber's omission of several recent texts on statistics in England. To this I would add my concern about the under-emphasis on the work that occurred before 1830. Malthus (who arguably provided the final intellectual push for the 1801 population census) is but one example; Condorcet, King, Halley, Petty, and Ricardo could be considered others.

Schweber instead chooses to begin her inquiry at the onset of what Hacking has called the "avalanche of printed numbers" and shows how this information was treated within the centres of higher learning. Although this is a reasonable starting point in many ways, it seems to me that the utility of demography and vital statistics was established before the discipline received institutional recognition. I feel that Schweber could have done more to remind readers that a good deal of the technical matter of demography existed long before 1830 and that disciplining statistics was therefore already underway.

These are minor issues, though, and do little to detract from the book's appeal, particularly as an account of a critical period in the history of demography and vital statistics. Schweber should be commended for her great attention to detail and for her contributions to understanding how and why disciplines emerge.