The only pair that has been counted

Exxon / Shell climate

Exxon scientists briefed management in 1977 that carbon dioxide from fossil fuels was the most likely way mankind was changing the climate. Shell’s 1988 greenhouse report treated the same fact as a planning constraint. In public, for years after, paid advertorials sold uncertainty. Supran and Oreskes later counted the two registers: about 80 percent of internal documents acknowledged human-caused warming; about 12 percent of the advertorials did. This room is that ratio, plus the two memos you should actually read. Climate Files keeps the PDFs. We keep the split.

Pair 1 of 4

Exxon 1977 management briefing

advertorial

Public

Exxon and later ExxonMobil public communications · 1989-2004

corporate advertising and issue ads

New York Times advertorials and similar

“Public ads stressed uncertainty: the role of greenhouse gases was unclear, more research was needed, and costly action could be premature.”

Private

James F. Black · 1977-07

Exxon Research scientist, briefing management

Internal Exxon presentation

“In the first place, there is general scientific agreement that the most likely manner in which mankind is influencing the global climate is through carbon dioxide release from the burning of fossil fuels.”

The briefing is a staff preference: the mechanism is known, the company should plan. The advertorial is an audience preference: the mechanism is still a debate. Same firm, two climates.

The tell · If the technical brief is older than the uncertainty ad, the uncertainty is assigned, not discovered.

How this became public: Inside Climate News archive; later counted by Supran and Oreskes.

Read this pair full-screen Next pair

Pair 2 of 4

Shell 1988 greenhouse report

advertorial

Public

Industry public affairs · circa 1989-1998

oil-sector communications of the late 1980s–90s

Advertorials and trade-group copy

“Public copy treated climate change as a theory under study, not as an operating constraint on the product.”

Private

Shell Groep · 1988

confidential staff report

The Greenhouse Effect

“The changes may be the greatest in recorded history… The report treats CO2 from fossil fuels as a real and serious risk, including sea-level rise and agricultural disruption, written for internal planning.”

Shell’s 1988 paper is a planning document. Planning is a private preference: treat the science as true enough to spend against. Advertising is the affordable one.

The tell · A confidential scenario that names dates and centimeters is the honest register.

How this became public: Published via Climate Files / Climate Investigations Center.

Read this pair full-screen Next pair

Pair 3 of 4

Advertorials that stress uncertainty

advertorial

Public

ExxonMobil · 1989-2004

paid New York Times advertorials

NYT advertorial series

“Advertorials told readers the science was uncertain, that climate models were unreliable, and that mandatory cuts could wreck the economy.”

Private

Exxon and Mobil scientists and planners · 1977-2003

internal reports and briefings

Internal research files

“Internal documents from the same decades treat anthropogenic warming as a serious, likely, and operationally relevant fact.”

This is preference falsification as a product line. The scientists were allowed a private climate. The public was sold a debate.

The tell · Count the modal verbs. Internals say will and is. Ads say could and if.

How this became public: Advertorials are public; internals surfaced via journalism and archives.

Read this pair full-screen Next pair

Pair 4 of 4

Eighty versus twelve

advertorial

Public

ExxonMobil advertorials (counted) · 1989-2004

public issue advertising

New York Times

“About 12 percent of ExxonMobil’s paid advertorials acknowledged that climate change is real and human-caused.”

Private

Supran & Oreskes, reading Exxon internals · 1977-2014

document analysis

Environmental Research Letters / follow-on work

“About 80 percent of Exxon’s internal documents acknowledged that climate change is real and human-caused.”

This is the rare pair that has been counted. Preference falsification is not a vibe. It is a ratio between two corpora.

The tell · When you can count the two registers, do. The model gets sharper.

How this became public: Peer-reviewed coding of already-public internals and ads.

Read this pair full-screen Next pair