Could the image of a four-drawer filing cabinet, whose drawers extend backwards into, well, near-infinity help explain some of society’s current communication disconnects? In a recent New Yorker article, Jill Lepore suggests you can divide all human knowledge into these four drawers: The little paper label on the top drawer says “Mysteries,” the second is “Facts,” the third is “Numbers,” and the bottom drawer is “Data.”

In her analogy, the Mysteries drawer (drawer 1) contains things only God knows, “like what happens when you’re dead.” In the past, this drawer would have been crowded with speculations on such matters as how distant are the stars, what happened to the dinosaurs, how do cells and molecules and atoms work? Thanks to advances in the sciences, these topics have been moved into the Facts drawer (drawer 2). That drawer “contains files about things humans can prove by way of observation, detection, and experiment.” The Numbers drawer (drawer 3) holds what you might think: censuses, polls, averages—stuff that can be counted.

It’s drawer 4 on the bottom, “Data,” that captures most of Lepore’s and society’s attention today. Humans cannot know data directly, in her metaphor, but must derive it from a computer. This drawer used to be empty but is now jammed full. More full than we can use with all practicality.

Not only do the drawers collect different types of knowledge and information, they work differently. They follow different logics. You learn about mysteries by revelation and the discipline that studies them is theology. You collect facts “to find the truth” and you study them by way of law, the humanities, and the natural sciences. Numbers are collected in the form of statistics, acquired through measurement, and you study them through the social sciences. Data analysis by computer enables prediction, pattern detection, based on data science.

For any complicated question (the example she uses is mass shootings in the United States), she says “your best bet is to riffle through all four of these drawers.” Each has something useful to contribute. However, the default in recent years has been to reach for that bottom drawer, as if data science contains the only answers. I saw evidence of the shortcomings of this approach in a news story last week about American students’ declining test scores in history and civics. One commentator noted that the data do not point to reasons for the decline. “Ongoing debates over how to teach history may well be getting in the way of actually doing it,” he said. Once the data are there, then what?

Data science certainly doesn’t preclude the need to open the other three drawers; nor does it demand that we renounce “all the other ways of knowing,” Lepore, a historian (drawer 2), says. Her article goes on to discuss other topics, but she also might have considered whether the main reason people today can’t seem to reconcile differing points of view is that they are basing their views on the contents of different drawers.

Another cultural columnist, Virginia Heffernan, writing in the current issue of Wired, pulls all this together in a way that emphasizes the importance of data science in an article about the complexities of manufacturing modern silicon chips, “I Saw the Face of God in a Semiconductor Factory.” She calls these chips “the engine of nearly all modern abstraction, from laws to concepts to cognition itself” (drawer 2). The global economy of semiconductor chips (drawer 3) is “as mind-boggling as cryptocurrency markets and derivative securities (drawer 4). Or as certain theologies, ones that feature nano-angels dancing on nano-pins” (drawer 1).

Another danger of over-reliance on technoscience and the hubris that goes with it is one familiar to people as far back as the ancient Greeks, whose myths addressed the world-changing intervention of fire. Just ask Prometheus how that worked out for him.

Further Reading
How Data Happened: A History from the Age of Reason to the Age of Algorithms by Chris Wiggins and Matthew L. Jones
Technologies of Speculation: The Limits of Knowledge in a Data-Driven Society, by Sun-ha Hong
“Frankenstein’s warning: the too-familiar hubris of today’s technoscience” by Richard King, The Guardian 30 Apr 2023.