Steven Shapin on the Trust Inherent in Science

The historian of science Steven Shapin, who has taught at Harvard since 2004, began his career as a geneticist, an experience he has drawn on across his work, perhaps nowhere more effectively than in his classic A Social History of Truth (University of Chicago Press). Hoping to demonstrate that, as he puts it, “Scientists, like the laity, hold the bulk of their knowledge, so to speak, by courtesy,” he begins by describing in technical detail some of his old laboratory work analyzing mammalian DNA:

Here was what I did. I was given some pieces of rat liver which I then minced and froze in liquid nitrogen; I ground the frozen tissue and suspended it in digestion buffer; I incubated the same at 50 degrees C for 16 hours in a tightly capped tube; I then extracted the sample with a solution of 25:24:1 phenol/chloroform/isoamyl alcohol and centrifuged it for 10 minutes at 1700 X g in a swinging bucket rotor.

And so on, for several more sentences impenetrable to the lay reader. Finally: “This was DNA; I had it in my hand; I had verified the facts of its composition.”

But how did he know? Only by relying on an in-principle limitless set of assumptions, each grounded in trust — one of Shapin’s keywords and the major subject of A Social History of Truth. “Of course, I could have ... adopted a skeptical posture about the truthfulness of the label on the ‘ethanol’ bottle, and consequently about the competence and honesty of whoever prepared the liquid.” But even if he tested the ethanol, he would only enmesh himself in further chains of faith. Scientific knowledge depends on these “trusting systems,” infinitely vaster than any given scientist or even any particular area of scientific inquiry.

Shapin’s major early works — Leviathan and the Air-Pump (Princeton University Press), co-authored with Simon Schaffer, as well as A Social History of Truth — blended sociology and history to tell a new story about early modern science, one in which the advancement of knowledge depends on reputational capital and the fundamentally social institution of credibility. “Cognitive order” and “moral order” are mutually interlocking. “What social conditions,” Shapin asks, “have to be satisfied for the collective good called knowledge to exist?”

Since the Covid-19 pandemic began, that question has become one to which almost everyone on the planet has some relationship. As Shapin wrote in the Los Angeles Review of Books in 2020, one of the “the biggest questions that the Covid crisis puts to us [is] whether we can recognize genuine expertise and act on it.” Nor are Shapin’s interests confined to the hard sciences. Like Michel Foucault or the philosopher Ian Hacking, Shapin has long pursued a concern, persistent if auxiliary to his main focus, with the way people’s sense of themselves is shaped by expert discourse.

Accordingly, Shapin, across his work, pays sometimes-glancing and sometimes-focused attention to psychology, phenomenology, taste, and perception — and, voluminously in the last two decades, to wine connoisseurship, a topic which, when we spoke recently by Zoom and by email, he did not neglect. Our conversation has been edited for length and clarity.

In the most recent History of Science, you published a history of the distinction between so-called “hard” and “soft” sciences.

That distinction doesn’t go back forever. It arose in post-World War II America, and, while it’s traveled around the world, it doesn’t sit well with non-Anglophone sensibilities about the human and natural sciences. I was interested in the hard/soft language because it contains both description and evaluation. It’s widely considered good to be hard, bad to be soft.

I called the distinction an “array” because it presumed the “soft” human sciences and the “hard” natural sciences to belong in one ordered series. People could think of sociology, say, as an inferior or undeveloped version of physics or chemistry. And, while there has been a certain amount of pushback, the human sciences — especially in this country — have labored to be “like” the natural sciences. Or at least to follow the methods ideally ascribed to the natural sciences. Not quite the same thing.