This essay argues that the explosion of visual graphics in nineteenth-century population statistics was closely linked to a shift in statistical epistemologies and practices of data collection. Taking German census statistics as a case in point, I illuminate concepts and practices that referred to data as a category of the here and now, enabling spatial representations of current phenomena. I argue that seeing and abstracting the world as data opened new avenues not only for producing tables with multiple variables, but also for forging such refined results into graphical visualizations of data. These in turn made empirical relationships in the social order evident and thus modifiable through intervention and reform. This essay is part of a special issue entitled Histories of Data and the Database edited by Soraya de Chadarevian and Theodore M. Porter.

The historiography of physics has reached a great degree of maturity and sophistication, providing many avenues to consider the making of science from a historical perspective. However, the big picture of the making of physics is characterized by a predominant narrative focused on a conception of disciplinary formation through leadership transfers in research among France, Germany, and Britain. This focus has provided the history of physics with a periodization, a geography, and a fundamental goal commonly considered to be conceptual and theoretical unification. In this paper, I suggest the interest of reassessing this picture by analyzing the temporal, national, and epistemological viewpoint from which it is written. I use for this purpose an exemplary case study: Adolphe Ganot’s physics textbooks in France and their translation by Edmund Atkinson in England. In this context, I suggest future avenues for the study of the making of physics as a discipline, which consider the canonical role of textbooks in disciplinary formation beyond the Kuhnian paradigm.