Colourless Writings of Statisticians and Their Distant Readers: Creating a New Mode of Reading in the Journal of the Statistical Society of London, 1838-1858

Established in 1834 for the collection of facts, the Statistical Society of London (SSL) played a central role in the moulding of statistical facts. The SSL defined statistical facts as the aggregation of numerous observations and reduced the value of single observations to isolated facts that alone could not be accepted as evidence. For the production of statistical facts, the SSL promoted two measures: conducting a coordinated observation to collect new facts and gleaning facts from existing literature. While the SSL left the former to governments, it devoted its resources to fostering the latter. To fulfil this mission, the Journal of the Statistical Society of London (JSSL) was created in 1838 as a virtual storehouse of existing facts where one could find facts of interest. The JSSL allowed individuals to share their small-scale observations for further aggregation as well as to publish statistical tables compiled from scattered facts that were already published elsewhere. As with its contemporaries, the JSSL was designed to serve posterity, which led the journal to include what apparently bore little importance at the time but might be of interest to readers in the distant future. The journal’s scientific missions resulted in the JSSL’s acceptance to publish colourless articles that provided no hypothesis, no conclusion and even no ‘original’ data. This paper explores how the new concept of facts in statistics shaped the practice of writing and reading among statisticians through the examination of the JSSL’s creation.