Agnotology

Robert N. Proctor and Londa Schiebinger · 2008

Thesis

Ignorance is not simply the absence of knowledge that research gradually fills in; it is often produced, by secrecy, by what researchers choose not to study, by knowledge that fails to travel, and by deliberate campaigns to create doubt, and it should be studied as carefully as knowledge itself.

The main ideas

Proctor names the study of how ignorance is made agnotology, a word the linguist Iain Boal coined at his request in 1992.
Ignorance takes three forms: a native state that research begins from, a lost realm that attention leaves out, and a strategic ploy built on purpose.
Industries manufacture doubt to delay regulation. A tobacco company memo of 1969 put it plainly: "Doubt is our product."
Governments remove knowledge through classification, which creates an unseen body of knowledge alongside the open one.
Some knowledge never crosses from one culture to another, because of who travels, whose interests it serves and what disciplines count.
Ignorance can be built into a whole social position, as in the philosopher Charles Mills's account of white ignorance.

A missing term

Robert Proctor, a historian of science, argues that the history and philosophy of science study how knowledge is made and largely ignore how ignorance is made. He asked the linguist Iain Boal for a word for that study; Boal coined agnotology in 1992, and it first appeared in print in a footnote in Proctor's Cancer Wars in 1995. This volume, edited with Londa Schiebinger and grown from a 2003 workshop at Pennsylvania State University, is the first collection devoted to it.

Proctor's introduction distinguishes three kinds of ignorance. The first is ignorance as a native state or resource: the starting point of inquiry, the not-yet-known that research works on. The second is ignorance as a lost realm or selective choice: what is left out because attention went elsewhere, sometimes by deliberate decision about what to study. The third is ignorance as a strategic ploy or active construct: knowledge withheld, or doubt created through misdirection and deceit.

The mechanism the taxonomy reveals is that only the first kind shrinks naturally as research proceeds. The other two are sustained by choices and interests, so more research does not remove them.

Implication: when a question seems unresolved, ask which kind of ignorance is at work. If it is chosen or constructed, the remedy is to change the incentives, not to wait for more data.

Doubt as a product

The volume's best-known theme is ignorance made on purpose. Proctor opens with a line from a 1969 memo by an executive of the tobacco company Brown & Williamson: "Doubt is our product." The industry did not need to prove smoking safe; it needed only to keep the question looking open.

Several chapters trace the same strategy in other fields. David Michaels, an epidemiologist, writes on manufactured uncertainty: how contested science is used to delay protection of the public's health and environment. Naomi Oreskes and Erik Conway show how climate science became a victim of the Cold War. David Magnus examines the debate over genetically engineered organisms, setting risk management against the precautionary principle. Jon Christensen looks at journalism's part in the "agnotology machine."

The mechanism is asymmetry. Establishing harm takes years of evidence; casting doubt takes funding for alternative explanations, demands for more certainty and access to the press. Every year of doubt is a year without regulation.

Implication: when a well-funded party argues that the science is not settled, ask who benefits from delay and what evidence would satisfy them. A standard of proof that can never be met is a tactic, not a scientific position.

Removing knowledge

Peter Galison's chapter, "Removing Knowledge: The Logic of Modern Censorship," addresses a form of ignorance produced by the state. Classification does not merely hide individual facts; it removes whole bodies of knowledge from public circulation, so that a body of classified knowledge exists alongside the open literature, developed by people who cannot discuss it publicly.

The mechanism is structural. Classified knowledge cannot be criticized, built on or corrected by outsiders, so it develops without the checks that open science relies on, and the open world remains ignorant of what has already been learned.

Implication: in fields where much is secret, such as defense, intelligence and some industrial research, the public record understates what is known. Treat the open literature as a partial map.

Knowledge that failed to travel

Some ignorance is not suppressed but simply never transmitted. Londa Schiebinger's chapter follows the peacock flower, a plant used in the West Indies to induce abortion. European naturalists brought the plant itself back to Europe, but knowledge of its use as an abortifacient did not make the voyage.

Schiebinger argues this was not a matter of suppression. The knowledge failed to travel, or to take root, for three reasons: colonial voyages of exploration were largely male; controlling fertility ran against the interests of empires that wanted larger populations; and disciplinary and professional divides determined what naturalists recorded.

Adrienne Mayor's chapter traces a related pattern in another field: indigenous knowledge of fossils in North America, from Claverack, New York, in 1705 to Agate Springs, Nebraska, in 2005, and its suppression in the scientific record. Alison Wylie applies agnotology to archaeology itself.

The mechanism is filtering. What crosses from one culture to another depends on who travels, what they are looking for and whose interests the knowledge serves.

Implication: when knowledge from another culture or field seems absent, consider that it may exist but was never recorded by the people who wrote the record.

Ignorance as a social position

The final part of the volume theorizes ignorance. Michael Smithson surveys social theories of ignorance. Nancy Tuana takes the female orgasm as a case study in the epistemology of ignorance, the study of how not-knowing is produced and maintained. The philosopher Charles Mills contributes "White Ignorance," his argument that ignorance can be structural, produced and sustained by a social position, so that members of a dominant group can fail to know things about their own society that its other members know well.

The mechanism Mills describes is that ignorance serves the position that holds it. What a group does not know can protect its interests and its view of itself, so the ignorance is maintained even when the knowledge is available.

Implication: ask what your own position makes it easy not to know. Ignorance that benefits the person who holds it is the hardest kind to see.

Limits

The field's descriptive and normative aims sit uneasily together. The philosopher Manuela Fernández Pinto argued in Social Studies of Science in 2015 that agnotology carries implicit judgments about which ignorance is wrong, and needs an explicit framework for making them.
As an edited volume, the book collects cases more than it builds a single theory. The philosopher Susan Dieleman, reviewing it alongside another collection on the epistemology of ignorance, argued that each book succeeds best at the task the other set itself.

Thinking, Fast and Slow — ignorance produced inside individual minds, by bias, rather than by institutions.
The Black Swan + Antifragile + Skin in the Game + The Bed of Procrustes — why people who bear no cost for being wrong can sustain error indefinitely.