In July 1977, a senior scientist named James Black stood before the Management Committee of the Exxon Corporation — the men who ran what was then the largest company in the world — and told them something that would, if acted upon, have reordered the next half-century of human history. There was, Black said, "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." A year later he refined the warning: doubling the carbon dioxide in the atmosphere would raise average global temperatures by two or three degrees Celsius — a figure that sits comfortably inside the scientific consensus today — and humanity might have only five to ten years before hard decisions became unavoidable.
Exxon listened. It built one of the most sophisticated private climate-research programs on earth. In August 1979 it outfitted a supertanker, the Esso Atlantic, with custom instruments to sample carbon dioxide in the air and ocean across the route from the Gulf of Mexico to the Persian Gulf. Its scientists published in peer-reviewed journals, modeled the warming, and reached conclusions indistinguishable from those of NASA and the academic mainstream. And then, beginning in the late 1980s, the same corporation spent the better part of three decades and tens of millions of dollars funding a campaign whose central product was the public impression that none of this was settled — that the science was uncertain, the models unreliable, the scientists alarmist, and the whole edifice possibly a hoax.
This node is about that campaign, the claim it manufactured, and the genuine scientific questions it cynically exploited. It is a delicate thing to map, because two contradictory facts are both true and both load-bearing. The first is that the organized denial of human-caused warming is not a grassroots skepticism but a documented, funded, and in places confessed public-relations operation, run with techniques lifted directly from the tobacco industry. The second is that beneath the manufactured doubt sits a real and legitimate scientific argument — not about whether the planet is warming or whether humans are the cause, where the evidence is overwhelming, but about how much, how fast, how bad, and what to do — and the tragedy of the denial machine is that by faking the first argument it poisoned the public's ability to have the second.
The greenhouse effect is not a late-twentieth-century theory invented to justify carbon taxes. It is nineteenth-century physics, worked out by people who had no idea it would ever be controversial. In 1824 the French mathematician Joseph Fourier described how the atmosphere traps heat that would otherwise radiate to space, keeping the Earth far warmer than sunlight alone would allow. In 1856 the American scientist Eunice Foote demonstrated experimentally that a cylinder of carbon dioxide heated more in sunlight than ordinary air, and speculated that more of the gas would mean a hotter planet. Three years later, working independently with far more precise instruments, the Irish physicist John Tyndall measured the infrared absorption of gases directly and identified carbon dioxide and water vapor as the atmosphere's principal heat-trappers.
In 1896 the Swedish chemist Svante Arrhenius did the arithmetic the others had only gestured at. Filling page after page with hand calculations, he estimated that doubling atmospheric carbon dioxide would warm the globe by several degrees — and noted, almost as an aside, that the coal-burning of industrial civilization was steadily adding the gas to the air. He thought this might be pleasant for Sweden. The point is that the basic mechanism — carbon dioxide is transparent to incoming sunlight but absorbs the infrared the warmed Earth radiates back, and more of it means a warmer surface — was established science before the automobile, before the airplane, before anyone alive today was born. It rests on the same quantum physics of molecular absorption that makes a microwave oven work.
What the nineteenth century could not supply was a measurement of whether the carbon dioxide was actually accumulating. That arrived in 1958, when Charles David Keeling began continuous measurements at the Mauna Loa Observatory in Hawaii, far from any local source. The resulting graph — the Keeling Curve — is one of the most consequential images in science: a line that has climbed without interruption from about 315 parts per million in 1958 to well over 420 today, higher than at any point in at least 800,000 years of ice-core record, its sawtooth seasonal wobble proving the planet itself is breathing the gas in and out while the baseline marches relentlessly upward. The carbon being added is identifiable as fossil in origin by its isotopic signature: it is depleted in carbon-14 and carbon-13 in exactly the way that ancient, plant-derived coal and oil are. This is not a model or a projection. It is a reading off an instrument, and it is the floor beneath everything that follows.
The phrase entered the language in 2015, when two investigations — one by the Los Angeles Times in partnership with the Columbia Journalism School, the other by the nonprofit newsroom InsideClimate News in an eight-month series titled "Exxon: The Road Not Taken" — reconstructed from internal documents the gap between what the company's scientists understood and what the company told the world. InsideClimate's reporters interviewed former Exxon employees and federal officials and read hundreds of pages of internal memos. The picture that emerged was not of a corporation that failed to grasp the problem. It was of a corporation that grasped it early, precisely, and in-house.
A 1982 internal report, prepared under Exxon scientist Marvin Glaser and stamped "not to be distributed externally," laid out the trajectory with what hindsight reveals as eerie accuracy, while warning that mitigation might "require major reductions in fossil fuel combustion." Two academics, Geoffrey Supran and Naomi Oreskes, later subjected the full corpus to formal content analysis. Their 2017 paper in Environmental Research Letters compared Exxon's internal and peer-reviewed documents against its paid "advertorials" — the op-ed-shaped advertisements it ran on the New York Times editorial page from 1989 to 2004 — and found a stark divergence: roughly 80 percent of the internal and academic documents acknowledged that climate change was real and human-caused, while roughly 80 percent of the public advertorials expressed doubt. In a 2023 paper in Science, Supran, Oreskes, and the physicist Stefan Rahmstorf went further, quantitatively grading the warming projections Exxon's own scientists had made between 1977 and 2003. The projections were not vague hand-waving. They forecast warming of about 0.20°C per decade, with a stated uncertainty of ±0.04°C — a number that has since proven not merely correct but, in the authors' assessment, "at least as skillful" as the contemporaneous projections of independent academic and government modelers. Exxon's scientists had, in the researchers' phrase, predicted global warming with "shocking skill and accuracy."
ExxonMobil has consistently rejected the characterization, arguing that its research was openly published, that selectively quoted documents misrepresent decades of evolving understanding, and that it has long supported a carbon tax and the Paris framework. That defense has force on the narrowest point — the company did publish — but it does not touch the core of the Supran–Oreskes finding, which is not about secrecy but about the documented contradiction between the science Exxon produced and the doubt it purchased. And Exxon did not act alone. It was one node in an apparatus.
The definitive account of that apparatus is the 2010 book Merchants of Doubt by the historians Naomi Oreskes and Erik Conway, and its central discovery is genealogical: the men who sold doubt about climate change were, in many cases, the very same men who had earlier sold doubt about tobacco, acid rain, the ozone hole, and secondhand smoke. A small group of politically conservative Cold War physicists — Frederick Seitz, a former president of the National Academy of Sciences who went on to consult for R.J. Reynolds Tobacco; S. Fred Singer; William Nierenberg; and Robert Jastrow, who together founded the George C. Marshall Institute in 1984 — moved from cause to cause, lending the prestige of their credentials to the project of keeping a scientific question "open" long after the scientists who actually studied it had closed it.
Their method was not to win the argument. It was to prevent the argument from ever being seen as won. And that method had a written charter. In 1969, an executive at the tobacco company Brown & Williamson had set it down with a candor that reads, in retrospect, like a confession for an entire industry: "Doubt is our product," the memo states, "since it is the best means of competing with the 'body of fact' that exists in the mind of the general public. It is also the means of establishing a controversy." You do not need to prove cigarettes are safe. You need only to make the public believe the question is unsettled — because an unsettled question does not compel action. The same logic, applied to carbon, would buy the fossil-fuel industry decades. This is the Invisible Control Systems playbook in its purest documented form: not the suppression of a fact but the manufacture of a fog around it, run through trusted-seeming experts and a press trained to give "both sides."
The receipts accumulated. In 1995 the Global Climate Coalition — an industry lobby including Exxon, Chevron, the American Petroleum Institute, and the auto and coal sectors — convened its own scientific advisory committee to prepare an internal primer. The primer's verdict, written by the coalition's own experts, was unambiguous: "The scientific basis for the Greenhouse Effect and the potential impact of human emissions of greenhouse gases such as CO2 on climate is well established and cannot be denied." The coalition's operating committee then ordered the section rebutting the contrarian arguments deleted before the document was circulated to members — a fact that surfaced only in 2009, when the draft was unearthed in litigation and reported by the New York Times. The lobby telling the public the science was uncertain had been privately told by its own scientists that the science could not be denied.
Three years later came the bluntest artifact of all. In 1998, in the wake of the Kyoto Protocol, a team convened by the American Petroleum Institute — including a representative from Exxon — drafted a "Global Climate Science Communications Plan," a multi-million-dollar, multi-year proposal that has come to be known as the "victory memo." Its metric for success was stated plainly: "Victory will be achieved when average citizens 'understand' (recognize) uncertainties in climate science; recognition of uncertainties becomes part of the 'conventional wisdom.'" Victory, in other words, was defined not as a true public understanding but as the successful installation of doubt as common sense. The plan called for recruiting scientists with no public profile on the issue, training them as media spokesmen, and seeding the impression of an open debate.
The political consultant Frank Luntz supplied the messaging discipline. In a 2002 strategy memo to Republican leaders, later leaked, Luntz wrote with the same clarity: "Should the public come to believe that the scientific issues are settled, their views about global warming will change accordingly. Therefore, you need to continue to make the lack of scientific certainty a primary issue in the debate." He added, tellingly, "The scientific debate is closing [against us] but not yet closed. There is still a window of opportunity to challenge the science" — and recommended retiring the frightening phrase "global warming" in favor of the milder "climate change." Luntz would later, to his credit, recant entirely, telling a Senate committee in 2019 that he had been wrong and now urged climate action. But by then the frame he had sharpened had done its generation-long work.
Behind the messaging stood money. The think tanks that produced the contrarian commentary — the Heartland Institute, the Cato Institute, the Competitive Enterprise Institute, dozens of others — were substantially funded by fossil-fuel interests, most prominently the petrochemical empire of Charles and David Koch through vehicles such as Americans for Prosperity. The sociologist Robert Brulle, in a 2014 study in Climatic Change, mapped ninety-one organizations comprising the climate "counter-movement" and estimated their combined income at roughly $900 million a year, increasingly routed through donor-advised funds like Donors Trust that obscured the original source. The Heartland Institute — which had earlier done work for Philip Morris questioning the health effects of secondhand smoke, completing the tobacco circle — erected a Chicago billboard in 2012 featuring the face of the Unabomber, Ted Kaczynski, beside the line "I still believe in Global Warming. Do you?" The backlash was severe enough that Heartland pulled it within a day, but the instinct it revealed — to brand acceptance of mainstream science as the mark of a lunatic — was the campaign's logic stripped of its lab coat. This is the Corporate Personhood & The Corporation thesis rendered atmospheric: a profit-seeking entity, having privately documented a planetary harm, deploys its capital to keep the cost external by purchasing public uncertainty about whether the harm exists at all.
Against the manufactured impression of a deadlocked science, researchers periodically tried to measure where scientists actually stood. In 2004 Oreskes published a short essay in Science reporting that of 928 peer-reviewed abstracts on climate change she had sampled, not one rejected the consensus position that human activity was warming the planet. In 2013 a team led by John Cook published the most-cited such study, in Environmental Research Letters: of 11,944 abstracts surveyed, about two-thirds took no position on the cause (as specialist papers on, say, Arctic ice mechanics often would not), but of the roughly one-third that did express a position, 97.1 percent endorsed the human-caused consensus. This is the origin of the famous "97 percent" figure, since corroborated by multiple independent methods and, in a 2021 follow-up, pushed above 99 percent in the recent literature.
The consensus is also institutional. Every major scientific academy on earth — the U.S. National Academy of Sciences, the Royal Society, and their counterparts in every G8 and most other nations — affirms it. The Intergovernmental Panel on Climate Change, whose reports are the negotiated synthesis of thousands of scientists and signed off line by line by member governments including the oil states, stated in its Sixth Assessment Report in 2021, in language the panel's consensus process makes deliberately conservative: "It is unequivocal that human influence has warmed the atmosphere, ocean and land." The word "unequivocal" is not used lightly in a document that hedges everything it can.
Here precision matters, because the "97 percent" figure is both real and routinely overstated in two opposite directions. Denialist commentary attacks it as fabricated, noting correctly that the Cook study's headline depends on excluding the no-position abstracts — but the exclusion is methodologically standard, since a paper that takes no stance is evidence of neither agreement nor dissent, and the result holds across every other survey that has tried. The opposite distortion, common in activist rhetoric, treats the 97 percent as if it certified every alarming prediction about sea-level rise, hurricanes, or policy timelines. It does not. What the consensus robustly establishes is narrow and foundational: the planet is warming, and human greenhouse-gas emissions are the dominant cause. It says far less about the questions that are genuinely contested at the research frontier — and those questions are where the honest skeptic actually lives.
Strip away the funded doubt and a residue of real scientific argument remains, and a fair map must hold it at full strength. The central genuine uncertainty has a name: equilibrium climate sensitivity, the amount of warming that results once the climate fully adjusts to a doubling of atmospheric carbon dioxide. For four decades after the 1979 Charney report, this number stubbornly refused to narrow, pinned at a "likely" range of 1.5 to 4.5°C — a span so wide it separates a manageable problem from a catastrophic one, and it sat there, barely improved, through generations of supercomputers. The reason is clouds. How a warming atmosphere changes cloud cover, and whether those changes amplify or dampen the warming, remains the single largest source of uncertainty in climate physics, and a low-sensitivity world is a scientifically defensible position, not a fringe one. Only with AR6 in 2021 did the assessed range tighten, to a best estimate of 3°C and a likely range of 2.5 to 4°C — and even that narrowing is contested in the literature.
The skeptic's case extends from there. Climate models are not data; they are imperfect representations that have, at various points, run warmer than observations, most visibly during the so-called "hiatus" in the rate of surface warming in the early 2000s, which the models did not anticipate and which took years to reconcile. The temperature reconstruction of the past millennium — Michael Mann's famous "hockey stick" graph, with its long flat handle and sharp recent blade — was the target of a serious statistical critique by Stephen McIntyre and Ross McKitrick, who argued its principal-components method could mine hockey-stick shapes from noise; a 2006 National Research Council review largely upheld the conclusion that recent warming is anomalous while conceding real uncertainty in the reconstruction before about 1600 and faulting some of the statistics. And the public face of climate advocacy has at times overreached in ways that handed the skeptics ammunition. Al Gore's 2006 film An Inconvenient Truth was found by a British High Court judge in 2007, in a case over its distribution to schools, to be "broadly accurate" but to contain nine claims that departed from the scientific mainstream — most notoriously the suggestion that sea levels could rise twenty feet "in the near future," when the IPCC's timescale for such a rise is centuries. Specific apocalyptic predictions with dates attached — ice-free Arctic summers by a given year, the disappearance of snow — have repeatedly come and gone unfulfilled, and each failure is harvested as evidence that the whole enterprise is alarmism.
The strongest version of the skeptical position, then, is not denial at all. It concedes that the planet is warming and that humans are the main cause, and contests instead the magnitude, the speed, the severity of the consequences, and above all the policy response — arguing that aggressive decarbonization may impose certain, concentrated costs on the poor today to forestall uncertain, diffuse harms tomorrow, and that adaptation and innovation may be wiser than rapid mandated transition. These are real arguments, addressed to the part of the question the consensus deliberately leaves open. They deserve a serious hearing. The catastrophe of the denial machine is that, by manufacturing a fake controversy about the settled part, it made it nearly impossible to have an honest controversy about the unsettled part — exactly the failure mode the Manufacturing Consent & The Propaganda Model model predicts, in which the structure of the debate is captured before any individual argument is made.
In November 2009, weeks before the Copenhagen climate summit, more than a thousand emails and documents were released — hacked or leaked — from the servers of the Climatic Research Unit at the University of East Anglia, one of the world's principal keepers of the global temperature record. The timing was surgical, and the contents were combustible. Two phrases became, for the denial movement, the smoking gun the whole edifice had promised.
The first was a 1999 line from the CRU's director, Phil Jones, referring to "Mike's Nature trick" to "hide the decline." Read cold, it sounds like a confession of fraud. Read in context, it is something far more mundane, and the distinction is the whole of the matter. "Mike's Nature trick" referred to a data-presentation technique used by the climatologist Michael Mann in a paper in Nature: a "trick" in the working scientist's harmless sense of a clever method, not a deception. "The decline" referred to the "divergence problem" — a well-known and openly published anomaly in which tree-ring data from certain northern sites, a proxy used to estimate past temperatures, inexplicably stop tracking the actual measured temperatures after about 1960. Because the instrumental record after 1960 is real and reliable, splicing it onto the proxy reconstruction for that period — rather than showing the proxy's known post-1960 failure — was a defensible way to present a reconstruction, and the divergence had been discussed in the literature for years. The "decline" being hidden was a decline in the reliability of certain tree rings, not a decline in global temperature.
Eight separate official inquiries on two continents reached the same verdict on the central charge. The UK House of Commons Science and Technology Committee, the Oxburgh Science Assessment Panel, and the Muir Russell Independent Climate Change Email Review in Britain; investigations by Pennsylvania State University, the Environmental Protection Agency, the Department of Commerce Inspector General, and the National Science Foundation in the United States — all found no evidence of scientific fraud, data fabrication, or manipulation of results. Lord Oxburgh's panel found "absolutely no evidence of any impropriety whatsoever." The science, every inquiry concluded, stood.
But honesty requires the other half. The inquiries did not give the CRU scientists a clean bill of health on conduct. The Muir Russell review specifically faulted them for "a consistent pattern of failing to display the proper degree of openness," for a defensive culture around Freedom of Information requests, and for treating critics as enemies to be stonewalled rather than answered. The emails showed real scientists behaving in real and unflattering ways — discussing how to keep skeptical papers out of journals, expressing the wish to delete data rather than release it to people they regarded as bad-faith antagonists, circling the wagons. None of that altered a single temperature reading or falsified a single conclusion, and subsequent independent reconstructions — including one led by the physicist Richard Muller, a former skeptic funded in part by the Kochs, whose Berkeley Earth project set out to check the record and confirmed it — vindicated the science completely. But the tribalism and the FOI evasiveness were real, and they were exactly the human failings most corrosive to public trust in a field that could not afford to lose it. The denial movement had manufactured a fraud that did not exist, and in doing so distracted from a genuine if lesser failing that did. Both things are true at once, which is why "Climategate" remains the affair where the careful cartographer must hold two readings in each hand.
The climate "hoax" claim, examined closely, dissolves into two very different propositions wearing one coat. The strong proposition — that anthropogenic warming is a fabrication, that the science is a fraud, that thousands of independent scientists across every nation and discipline are either deceived or in on a conspiracy — is not supported by any evidence and is contradicted by the basic radiative physics, the instrumental record, the isotopic fingerprint of the added carbon, and the documents in which the fossil-fuel industry's own scientists confirmed all of it before their employers paid to deny it. The weak proposition — that the magnitude is uncertain, that models have limits, that predictions have failed, that policy involves real and contested tradeoffs, and that climate advocacy has at times been alarmist and politicized — is largely true, and is the legitimate domain of scientific and democratic disagreement.
The denial machine's lasting achievement was to weld the two together so that the weak, defensible proposition became the camouflage for the strong, indefensible one. Every real uncertainty about clouds and sensitivity, every overreaching prediction, every unflattering email was deployed not to refine the science but to suggest the whole structure was rotten. This is why the phenomenon belongs in a map of contested power rather than a map of contested physics. It is a worked instance of the Invisible Control Systems thesis — manufactured perception substituting for force — financed by the Corporate Personhood & The Corporation form's structural drive to externalize cost, and it is fought, on the policy flank, on the same ground as the Agenda 21 & Sustainable Development and The Club of Rome & The Limits to Growth anxieties, where the suspicion that planetary problems are pretexts for planetary government gives the doubt a political home.
And that suspicion is the hardest thing to dismiss, because it is not entirely baseless. The fear that "climate" can be made the all-purpose justification for control — for carbon budgets, movement restrictions, programmable money, and the transfer of decisions from elected legislatures to unaccountable expert bodies — is a coherent worry that does not require the science to be false. It is possible, in principle, for the warming to be real and for some responses to it to be authoritarian overreach; the two questions are logically separate, and conflating them is the error committed on both sides. The denialist conflates them to discredit the science by association with the policy; the alarmist conflates them to discredit the policy critique by association with denial. The merchants of doubt understood this perfectly. Their genius was never to deny the warming outright — that was always too easily refuted. It was to keep the public permanently unsure, so that the cost of acting always seemed to exceed the cost of waiting, for just long enough that the waiting itself became the most consequential decision anyone ever made.