More than four decades ago, the Italian historian and critic Carlo Ginzburg argued that the modern disciplines of knowledge that had arisen in the late nineteenth century relied on the interpretation of clues. With medicine—symptomatology—as the paradigm, these disciplines were concerned with deciphering signs, wrangling indications from seemingly mute traces. The target of the interpretation was individual persons and behaviors, as the example of criminology showed. The means was hypothesis, a sort of divination with an “inevitable margin of hazardousness.”11xCarlo Ginzburg, “Clues: Roots of a Scientific Paradigm,” Theory and Society 7 (1979): 281. The result was social control—empirical knowledge of even unintended behaviors allowed for prediction and correction. Clues, Ginzburg concluded, had become paradigmatic for the sciences in the 1870s and 1880s. And the most proximate complement to medicine in this regard was philology. Criminology and literary interpretation had a logic in common. Today we would call it the logic of data.
Alan Turing never used the word data in the 1936 paper that defined computation and launched us into the digital world we live in today.22xAlan M. Turing, “On Computable Numbers, with an Application to the Entscheidungsproblem,” Proceedings of the London Mathematical Society, 2 (published 1937, written 1936), 42 (1): 230–65; https://www.csee.umbc.edu/courses/471/papers/turing.pdf. The term appears occasionally in the work of Turing’s peer John von Neumann, after whom modern computer architecture is named. The term, which already referred to information stored on paper, the results of bureaucratic labor, seeped slowly into computer discourse in the 1940s and ’50s, even as data became literal inputs into room-sized machines like the ENIAC, entered on punched cards. Seven decades later, those input numbers have gained a life of their own. They swirl around us, only sometimes touching down long enough for us to make any sense of them. We use these numbers as signs to navigate the world, relying on them to tell us where traffic is worst and what things cost. And because we do this, data has become a crucial part of our infrastructure, enabling commercial and social interactions. Rather than just tell us about the world, data acts in the world.
The Latin meaning of data is “givens,” but data in its modern meaning refers not to gifts of nature but to the input and the output—endlessly feeding back into one another—of digital machines. These machines send us messages—push notifications in the form of little hermeneutic puzzles, signs to read off screens. Data is both representation and infrastructure, sign and system. Think of the just-in-time logistics of Amazon’s delivery game. You click on a few icons to complete a purchase, and a series of events begins—involving robots, deplorably underpaid and overworked laborers, and parcel tracking. Data was the channel along which the prices were set and the items offered to you as icons on your screen. But it’s also the channel in which all the supply-side decisions are made, often automatically. Warehouse stocks and delivery routes change, and so do prices. Data makes all of this possible, but it is also the medium in which it is carried out—as media theorist Wendy Chun puts it, data “puts in place the world it discovers.”33xWendy Chun, “Queerying Homophily,” in Pattern Discrimination, ed. Clemens Apprich et al. (Minneapolis, MN: University of Minnesota, 2019), 62. Even the labor is done at the command of data, which both represents and determines the process. The numbers Turing put into the machine have become an array of signs about the world that also act in the world. We read them and act according to them; algorithms predict and influence our behaviors by means of indexes wrung from data.