The mechanism is seventy years old and operates with the precision of a Swiss watch. It does not consist in denying science, but in multiplying it until it becomes unreadable. It does not suppress facts; it surrounds them with so many alternative questions that they cease to be facts and become opinions. What is today called scientific disinformation has a precise date of birth, a place, names, and ninety-three million pages of evidence that remained under lock and key for decades in archive warehouses. This is the story of how industry turned doubt into its most profitable product, and of the academic discipline that took half a century to find a name for what had happened.
The hotel, the men and the memo
In December 1953, the presidents of America’s seven largest tobacco companies met at the Plaza Hotel in New York. Outside, the scientific studies were accumulating. The relationship between tobacco and lung cancer was, according to researchers of the time, overwhelmingly clear. For the companies of what is known as Big Tobacco, the problem was not science itself, but that they could not refute it. Their own internal teams knew this and documented it. What that meeting produced was not a scientific denial, but something more sophisticated and more enduring. A strategy to make certainty appear provisional.
The decision taken that evening was, on the surface, virtuous. The tobacco companies announced the creation of the Tobacco Industry Research Committee and pledged to fund science on a massive scale. The press received it as a gesture of corporate responsibility. No one saw, because no one could see, that the science they were about to fund had a specific function and that function was not to find the truth. The research produced by that committee focused in the following years on radon, asbestos, viruses, personal habits, and even the month of birth as a cancer risk variable. Each of those factors exists, has real effects on health, and was studied with sufficient methodological rigour to be published in reputable academic journals. That was precisely the point. The aim was not to produce lies, but to produce such a quantity of partial truths that the central truth would be buried beneath them.
The historian of science Robert Proctor gave this process a precise name. Agnotology, from the classical Greek for ignorance. Not the study of what we do not know, but the study of how we actively produce what we do not know. The distinction matters. For centuries, ignorance was conceived as a natural state, the starting point from which science progressively moves us away. What the tobacco case revealed is that ignorance can be manufactured, funded, sustained and monetised like any other industrial product. Agnotology is, in that sense, the discipline that studies science’s internal enemy.
The factory exports its model
What followed in the subsequent years was neither an exception nor an anomaly. It was a technology transfer. The model built by the tobacco companies was studied, adapted and replicated by industries facing the same structural problem. Scientific evidence was accumulating against their products, regulations were threatening their markets, and legal action was on the horizon.
The case of the bees illustrates the mechanism with clinical precision. From the 1990s onwards, beekeepers across Europe and North America were documenting massive colony mortality. Entomologists were converging on a probable cause. Neonicotinoids, a new generation of systemic insecticides, that is to say insecticides integrated into the plant tissue itself, were being consumed by bees as they fed on nectar. The evidence was not anecdotal; it was the result of thousands of field and laboratory analyses. What followed was predictable for anyone who knew the history of tobacco. Studies proliferated on the varroa mite, on nosema disease, on the Asian hornet, on climate change, on habitat loss. Each of these factors genuinely contributes to the decline of bees. None of them alone explained the geographical synchrony of the mass die-offs. But together they produced what the industries needed. A context of sufficient uncertainty to paralyse regulation for two decades.
The biologist Frederick vom Saal published in 1997 the results of his research into bisphenol A, a chemical compound found in everyday plastics. His experiments on mice revealed effects on the reproductive system at doses twenty-five thousand times lower than those that regulatory toxicology considered safe. A subsequent analysis of the available scientific literature found that 93 per cent of research funded by public sources confirmed effects of bisphenol A at low doses, compared to 0 per cent of studies funded by the plastics industry.
Bisphenol A added a layer of complexity to the model. What vom Saal discovered in his laboratory was not only that the product was toxic, but that it was toxic in a way that conventional toxicology was not designed to detect. The classical rule of toxicology is proportional. The higher the dose, the greater the effect. Bisphenol A acts as an endocrine disruptor, that is to say a hormone mimic, and its effect is more pronounced at low doses than at high ones. This means that the standard risk assessment protocols used by regulatory agencies across the world to approve or prohibit substances were measuring in the wrong range. The industry did not need to falsify the data. It was sufficient to defend the existing protocols.
The archives that changed history
For forty years, the strategy operated in silence. What broke it was neither a journalistic investigation nor a government report. It was an unmarked box of archive documents. In May 1994, the researcher Stanley Glantz, at the University of California in San Francisco, received by post a box containing internal documents from the tobacco companies. More boxes would follow, smuggled out clandestinely by an employee who slipped the materials under his shirt past the security guard and photocopied them before returning them to their place. When the American courts intervened and compelled the tobacco companies to open their complete archives, the resulting volume was of a scale no researcher had anticipated.
Among the ninety-three million internal documents from the tobacco industry, made available to the public by court order in the late 1990s and archived at the University of California in San Francisco, is a memorandum from Brown & Williamson dated 1969. “Doubt is our product, 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.”
What those ninety-three million pages document is not merely the corruption of one industry. It is the systematisation of a method. The archives contain contracts with scientists recruited to produce diversionary research, minutes from meetings where strategies were designed to influence epidemiological evaluation protocols, budgets allocated to networks of spokespeople dispatched to academic conferences and television programmes, and the transnational coordination between Philip Morris in Europe, British American Tobacco in Asia, and Reynolds in the Pacific. The scale was global. The coordination was explicit. And it was all written on the companies’ own headed notepaper.
Naomi Oreskes, historian of science at Harvard, used those archives to trace the full lineage of what she and Erik Conway called the merchants of doubt. Her work demonstrated that several of the scientists who led the denial of climate change in the 1980s and 1990s had previously participated in the tobacco companies’ campaigns. It was not a coincidence. It was the same network, with the same arguments and the same techniques, applied to a new adversary. You do not rewrite a manual that works.
Ideology as an autonomous engine
The physicists who built the initial academic resistance to the climate consensus were not, for the most part, paid by the oil companies. They were men formed during the Cold War, who had devoted their careers to defending American technological supremacy against the Soviet bloc. Fred Singer, one of the pioneers of the American space programme, spent the 1990s travelling between conferences to cast doubt on climate models with the same formal rigour he had applied for decades to rocket physics. When the communist world collapsed and the environmental sciences began to demand state regulation of the market economy, these men saw in that movement precisely what they had fought against for decades. A collectivist threat to the free market, now clothed in ecological language rather than Marxist language. Their resistance was sincere. This made it more durable than any commissioned campaign, and considerably harder to dismantle.
The Heartland Institute, a think tank based in Chicago dedicated to promoting the free market, is the institutional heir to that tradition. It is not a scientific organisation, though it produces documents that mimic the format of academic output. Its representatives travel to conferences, publish in conservative media and manage digital networks whose activity David Chavalarias’s team at the Paris Institute of Complex Systems described in a 2016 study as structurally hyperactive. A small core of accounts compensates for its numerical scarcity with a publication frequency that saturates algorithms and colonises spaces of debate. They are fewer in number. They publish far more.
What agnotology reveals, in its most uncomfortable version, is that manufactured ignorance does not always require a corporate patron. It sometimes sustains itself, fed by genuine convictions that coincide, for structural reasons, with the interests of those who have something to lose. Industry learned early to distinguish between the denialists it finances and those who work for free. The latter are more effective. They are also more difficult to identify.
The method survives the scandal
The scandal should have ended with the release of the archives. The documents have been public for decades. The strategy is described with clinical precision in thousands of pages signed by the executives themselves. The names are in the contracts. And yet, the industries that inherited the Plaza Hotel manual are operating this week with the same techniques, on new products, before new regulatory agencies, in new countries. Tobacco taught the world that scandal and consequence are two different things. That the exposure of a method is not equivalent to its deactivation. That the doubt factory does not close when investigated. It simply hires better lawyers, changes its headed notepaper, and continues to open for business.
Agnotology was born from the conviction that naming a mechanism was the first step towards dismantling it. That is a reasonable wager and, up to a point, a true one. What ninety-three million pages demonstrated, without intending to, is that there are systems designed to survive their own exposure. Doubt as a product does not need to remain hidden in order to function. It can operate in full daylight, with the archives open and the strategy published in university textbooks, because the objective was never secrecy. The objective was time. And time, across seventy years of documented history, has always vindicated the industry…
G.S.
Sources
- Naomi Oreskes and Erik Conway, Merchants of Doubt (Bloomsbury Press, 2010)
- Robert N. Proctor, Golden Holocaust: Origins of the Cigarette Catastrophe and the Case for Abolition (University of California Press, 2011)
- UCSF Industry Documents Library, Tobacco Archives
- Frederick S. vom Saal et al., “A physiologically based approach to the study of bisphenol A”, Environmental Health Perspectives, 1998
- David Chavalarias et al., study on climate-sceptic networks, Institut des Systèmes Complexes de Paris, 2016
- Brown & Williamson, “Smoking and Health Proposal”, 1969, UCSF Truth Tobacco Industry Documents
- Heartland Institute, Climate publications
