The Misinformation Age by O’Connor and Weatherall

Cailin O’Connor and James Owen Weatherall. The Misinformation Age: How False Beliefs Spread. New Haven: Yale University Press, 2019.
Summarized by John Teevan, February 2019

O’Connor and Weatherall make the main point that reason and research, while objective and spreading to other scientists through reliable networks, are subject to propagandists and social networks.

Ch 1: What is truth?

The hole in the Arctic Vortex was discovered by measuring the frigid temperatures where the hole formed. Initially, satellites did not confirm the results, but later scientists noted that some satellite data were discarded as outliers. Nations responded to the CFC cause of this hole, and the danger has abated. Good science was quickly trusted and widely spread, leading to appropriate action.

Our authors acknowledge Pilate’s question, “What is truth?” a question behind facts acknowledged as at least a bit uncertain due to the problem of Induction (27). From a pragmatic perspective, however, absolute certainty is not necessary, though people (industries) have their own reasons to cast doubt. This was the main opposition to taking anti-CFC action to protect the ozone layer. It’s more important to think in terms of degrees of credibility, as in Bayes’ Rule: certainty increases with the number of researchers who confirm. As with the satellite-outlier issue, people interpret research through their existing paradigms. The change of paradigms is as essential as changing paradigms is difficult.

The next paradigm shift that was attacked concerned the problem of acid rain in Europe following an 1783 eruption in Iceland. This issue was clearly affected by political concerns. It set off what was called the science wars based on the idea that leftists were using this crisis for political power and ideology.

Ch 2. Polarization and Conformity.

This long chapter begins with a question about the dangers of mercury in fish. Doctors began looking for symptoms of mercury poisoning, then 20/20 ran a story, and fish sales fell. We will change our beliefs in the light of new evidence more readily if we have confidence and trust in the source. Lacking trust is a significant problem. But we don’t face that when we investigate on our own (52-3). Here, they introduce the chart of relationships between people represented by dots. There is a bias towards conformity in addition to the preference for solid evidence. Conformity to a trusted source helps spread innovative ideas, but it also can spread false beliefs.

Stomach ulcers are an example: people wrongly thought antibiotics were a ridiculous treatment. There needs to be a period of temporary diversity before good innovative ideas can gain full acceptance.

Lyme Disease (64ff): For decades, there have been two views of Lyme’s disease that have led to almost violent disagreement among doctors. Antibiotics either help or harm patients. In fact, one side is harming patients, but no one knows which side of that is. Our authors give political examples from Steve Scalise and the Charlottesville protests. As with Lyme’s disease, the polarization has become extreme.

As a base model, think of two modules, each with six well-networked members. These two independent modules can easily hold substantially opposing views, with some taking it personally (73). Now we have moved from a consensus model to a confirmation bias, but there is a third model. Sadly, when there are two opposing networks, there is extreme pushback. Jeffrey’s Rule allows for an erasure of uncertain/contradictory evidence. Since there is one best way, mistrusting authoritative people is toxic, as they might have the answers.

Semmelweis, an MD in Vienna, noted that many more newborns were dying in one hospital than another. Pregnant mothers were distraught. Semmelweis thought failure of handwashing after touching corpses led to the higher death rate at one hospital. The doctors, claiming they were gentlemen, could not imagine that they were being considered unclean only because they had not washed their hands. These docs hounded Semmelweis, who was arrested, beaten, and died in a mental hospital at the age of 42. His correct belief was despised…until it was accepted.

In the 1950s, Solomon Asch showed people four lines and asked which of the last three lines was equal to the first. Using individuals and groups, he included people who had been instructed to select the wrong answer. Many were so inclined to accept others’ beliefs that they nervously chose the wrong line to fit in. This error spreads through what they call “cascading information.” This is the social bias that our authors are focusing on. Call it conformity bias: we are inclined to distrust ourselves to fit in with others.

At a basic level (not the full networks of modules in the scientific community), there can be only two modules with only a couple of points of connection between them (87). Will reasonable people listen to those from another module? There is a trade-off between the desire to conform and the benefits of agreeing with your own research. These notes appear in areas such as resisting vaccination (Ch 4).

Ch 3. Evangelization of the Peoples:

In 1953, Reader’s Digest published an article, “Cancer by the Carton,” which was the first popularization of cancer research, showing a tenfold increase in lung cancer incidence among smokers between 1920 and 1948. Lung cancer is preventable, they said, if people would quit smoking, and many did. Immediately, the tobacco industry launched a marketing campaign to push back, not with falsely contrived studies, but with alternate studies that were less certain about tobacco’s role in cancer.

The tobacco evangelists/propagandists formed the credible-sounding Tobacco Industry Research Committee to make credible statements that were nevertheless a mass manipulation of public opinion. (98). They used surveys of subjective opinions, and they used smoking Lucky Strike-paid women at the Easter Sunday parade in New York to create distractions from the science.

“Those who manipulate the unseen mechanism set of society constitute an invisible government which is the true ruling power of our country,” said Bernays 99.

The tobacco strategy was wildly successful until 1970 and then 1992, when science prevailed. Using networks again (103) shows not only how consensus forms within a network of scientists but also how it can spread to policymakers who observe their work. If the policymaker listens to only one or two sources, they may reach a different conclusion than the scientists who use their full networks. Behind the policymaker stands a propagandist who helps them select which scientists to listen to. This is the theme of the chapter; there is an accusation of false science, but only of selecting scientific studies that sincerely reach conclusions more in line with the propagandist. This cherry picking includes leaving out contrary or ambiguous conclusions. It assumes the actual studies will never be read. Many module diagrams here demonstrate how good studies can be used in slanted ways.

Without interfering with the scientific process, they emphasize studies favorable to their industry and share only those with policymakers. More studies confirming results are not always funded due to limited funding. Still, that lack of replication undermines the proper conclusions of good studies and prevents refuting poor studies that cannot be replicated.

The coal industry has funded its own studies, raising questions about bias toward the one paying the researchers’ bills, a concern another study has confirmed (119). Studies include models that fit the facts, and yet propagandists seek models favorable to their views, and, therefore, favor the scientists who utilize these models, though they are not corrupt themselves.

Another tool of the propagandist is to measure the wrong thing. When a-fibrillation was considered an indicator of an approaching heart attack, there was a choice to measure the effect of a medicine by measuring whether it reduced a-fib or if it reduced deaths. (122). By focusing on a-fib, they missed the increase in deaths of those taking the med; hundreds of thousands of people died prematurely.

Using another tool, the propagandist recruits a ‘star’ who is at the center of a network of scientists. His credibility co-opts the trust factor and multiplies confidence in his work among those who trust him.

Turning to the market, the propagandist can use emotion, as with the Marlboro Man, to create an image that conveys trust and authority, which play a crucial role in shaping consumers’ actions (126). During the 1920s, a desire to sell bacon led to the invention of the notion of an American Breakfast of bacon and eggs. Bernays even found a physician who said a hearty breakfast, including bacon, was healthier. Though that was not a scientific conclusion, he was at least trustworthy.

Roger Revelle studied the ocean during and after WW2 for the Navy and later published an essential article concerning the rate at which carbon dioxide is absorbed into the sea (128). This is the greenhouse gas study that led to concerns about climate change. While greenhouse gases are essential to keeping our planet above 0°, an excess of carbon dioxide significantly raises air temperatures. While the ocean absorbs carbon dioxide, its capacity to do so decreases as CO2 levels rise. In 1965, Revelle moved to Harvard and met a young undergraduate, Al Gore, who gave Revelle a strong voice.

Those opposing Revelle found a partner for him in his old age, and together their article concluded “the scientific basis of greenhouse warming is too uncertain to justify drastic action at this time “(131). Revelle may not have said that, but he did not get a chance either way, as he died following heart surgery in July 1991, just after the article was released. His authoritative name on the article undid the impact of his earlier work and left Al Gore explicitly refuted.

Our authors call this a weaponization of reputation. Once we choose to trust someone, we continue to trust them. Now there are two climate change national organizations with competing views, each with its own aura of authority.

Finally, Lady Mary Montagu, a British aristocratic beauty in the 1690s (and friend of John Locke), was marred by smallpox as an adult and was therefore intrigued by the vaccination she saw in Turkey. She observed doctors scratching bits of smallpox into people’s arms. As a result, she became a star opinion leader in Britain, helping to speed vaccinations that have had a profound effect on public health. Nevertheless, a tight-knit Molly American community in Minnesota experienced an outbreak of measles because people do not believe in vaccinations. These people who believe that vaccinations cause autism have allowed a conquered disease to reestablish itself. Anti-vaxxers use their networks so that peer pressure to conform multiplies those who resist vaccinations. This tight group (diagram on 145) relies not on trust but on peer pressure to reach disastrous, untrue conclusions.

Ch 4. Social Networks.

The problems of scientists reaching an accurate consensus (Ch 2) and the misuse of scientists’ networks for propaganda (Ch 3) ultimately lead to the book’s essential topic: the impact of social networks. Our authors start with the 2016 story of a Mr. Welch who left his home in North Carolina to shoot up a pizzeria in Washington, which he believed was involved with a Clinton-Comey conspiracy to enslave children. He did so, based on fake news concerning an earlier raid.

The story was carried not only by the spurious Alex Jones, but also by the NY Times and the Washington Post, which published articles refuting it. Unlike scientists who want to discover truth or propagandists who wish to to shade its implications or cause confusion, social networks are freelance (or vulgar democratic) networks that combine malice and ignorance to spread gossip and fake news, sometimes intentionally.

Sharply divided political opinions over Obamacare, Iran, gun control, and tax cuts were dealt with in highly opinionated but legitimate networks. Even though there is an open bias toward their own perspective, they are not as irrational or outrageously emotional as social networks, let alone as sober as scientists.

Fake news is far from new. In 1898, the New York World blamed Spain for sinking the USS Maine in Havana harbor and harangued the U.S. to invade Cuba. Edgar Allan Poe had his own fake news about a warm-air balloon journey. The authors cite still another howler about a trip to the moon.

Facebook has 1.6 billion users and Twitter about half that many. Bogus stories may have always existed, but now they can spread as never before (154). TV coverage gives credibility and makes real and fake stories indistinguishable.

What are some problems of new coverage today? 1) Journalists seek the novel; they seek the outlier; they seek the interesting; and they seek the controversial. 2) The government had a TV Fairness Doctrine requiring reports on both sides of an issue, even if one was a severely minority view or even bogus. 3) Sometimes the facts are speculative on both sides of an argument, as with measuring weapons of mass destruction, making it difficult to tell who has the fake news. Our authors recommend trusting the Bureau of Labor Statistics, the International Panel on Climate Change (IPCC), and the UN.

In July 2016, Seth Rich was shot dead in Washington, DC. The murder was a mystery, but his work for the Clinton campaign led to rumors that the Clintons had killed him for threatening to expose their election fraud concerning Bernie Sanders. Related stories had previously accused the 1990s Bill Clinton campaign of two dozen murders. The Seth Rich story was also connected to the 2016 Russian hack of DNC documents, embarrassing to Hillary Clinton. The hacker Guccifer 2.0 and the WikiLeaks hacker Assange both entered by connecting their stories to real people and issues. Rich, however, was not involved, and Fox News retracted its statements. In the story, actual facts are mixed with fake news, making it easier for the misinformation to spread through social networks. The lack of a gatekeeper (168) is a problem. “Stories in which every sentence is true and impeccably sourced. It can nonetheless contribute to fake news and false beliefs. Our authors suggest not adding factual dimensions to fake news stories.

Our authors want to discuss Russian influence in the 2016 campaign, but limit themselves to how Russians can gain credibility with a variety of affinity groups, ranging from Black Lives Matter to animal lovers. Once Russians gain credibility, they promote their own propaganda. A second strategy is to create a Star (the center of a Star Network, Ch. 3). This, as with scientists, is designed to build trust and then to use trust for propaganda; a corruption of using it for scientific consensus, as in chapter 2.

Our authors recommend that social networking sites screen and remove fake news, as controlling its spread is their responsibility. Again, they call for editorial discretion (174). Goodhart’s Law states that when a measure becomes a target, it ceases to be a good measure because targeting leads to gaming and manipulation. The example of the French paying a bounty for rats’ tails in the early 1900s failed to account for the fact that the rats had not been killed. His idea demonstrates that control, algorithms, or editors can only temporarily stop fake news as it will seek to game new avenues.

Our authors close with nine recommendations:

1. Keep the news as local as possible to maximize accountability.
2. Construct social networks that minimize an entirely homogeneous subgroup of ‘true believers.’ “In an ideal world, politicians might play this role” (178). At least, they believe, politicians can be credible.
3. Quote a maverick, such as John McCain, who breaks from his group to agree with the other side, is more credible than any spokesman already committed to a given position.
4. The marketplace of ideas, while endorsed by Oliver Wendell Holmes in terms of free speech, nevertheless leads to social damage and needs to be rethought. They oppose free speech.
5. Scientific communities must adopt norms of publication that are even further peer reviewed so that minority opinions and dissenting opinions are removed and not likely to cause public confusion. The government or the National Academy of Science can oversee this aggregation
6. We must abandon industry funding of research. It isn’t very objective. (See JT response).
7. Journalists need an innovative approach to reporting about science to avoid the novel and sensational.
8. The U.S. should follow Europe’s lead in legal action to suppress fake news. The authors realize the risk to free speech but consider widely believed lies to be even more damaging.
9. Reimagine democracy so that majority rule does not evolve into a vulgar democracy as a result of ignorance (185), the Phil Kitcher who prefers the use of expert opinion alone in matters of fact, science, and policy. Kitcher immediately acknowledges that scientists are ignorant of people’s lives, even as people are ignorant of science, so he prefers a well-ordered science that would align with society’s ideals, though our authors question his utopianism and recognize its impracticality. Yet they endorse a new, practical, and dynamic form of Kitcherism. (END)

JT Response

I needed to read this book as I feared I was becoming too distant from the facts and too rigid in my opinions. This helped quite a bit. Here are my concerns:

1. The chapters cover varying degrees of distortion of the key players: the scientific community, propagandists, journalists/TV, and the social network. I believe they have left out the politicians as key players who distort and manipulate facts to advance their own re-election and ideology. For example, I’m not sure why the authors didn’t warn us about the dangers of government funding of research, because the government will not fund research that doesn’t meet minimum ideological standards.
2. Anti-vaxxers have always been a concern, but we have reason to be concerned about people who believe, against all science, that generating electricity from nuclear is unacceptably dangerous? Or people who believe, against all science, that genetically modified (GMO) crops and food from those crops are dangerous? These are superstitions. France loves nuclear energy while fearing GMO, and we fear nuclear, while, till recently, imitating the Europeans, accepting GMO foods along with the rest of the world that is grateful for GMO.
3. When politicians get involved, there is a discussion of political topics that trump science. Many strongly believe, for reasons of the rights of women, but against all science, that a full-term fetus is not a viable human being. How has the medical profession been silenced on this topic, and how have some doctors embraced the idea that even a breathing newborn can still be left to die for any reason that would have been acceptable for an abortion an hour earlier? Don’t care about dead babies? Why not care about the thousands of birds killed by politically favored wind turbines?
4. Their final recommendation that we eliminate ‘vulgar democracy’ is clear bias not only to government institutions but to the elites who, like our authors, believe that they are fully competent to set the standards that will rule the lives of ordinary people. Institutional bias and a bias toward elites are exactly the biases that government leaders and elites cannot see, though they lecture ordinary folks about our biases.
5. On the specific topic of climate change, I do not count myself among the ‘deniers.’ But so, what? They assume that once we acknowledge the dangers of climate change, the steps to take will be obvious. I have no confidence at all in politicians taking these ‘obvious’ steps. Maybe a carbon tax is a clever idea. Still, outlawing the internal combustion engine in 1970 with a five-year deadline (California 1970) or PM Merkel’s decree to end the nuclear power plants already working in Germany (2015) are much closer to absurd than to obvious.

The European Union, for example, decided to help save the planet by switching automobiles from gasoline to diesel. This not only failed but also added sulfur to the atmosphere and created bizarre incentives, including leading Volkswagen to adjust its vehicles so that VW engines gave false (but legal) readings during emissions testing. Now VW faces a billion Euro fine for failing to adjust to a failed government cure for climate change. I’m struggling to find an apparent reason to trust governments to deliver a proper response to climate change. Are the Chinese communists better? Not with all their current elevated levels of pollution, and with the development of as much new coal-fired electricity production as the U.S. currently has at full capacity.

I’m wondering who the wildly irrational and incompetent people are here. Exactly who are the people who are suppressing the true facts in favor of fake ones? Are we really going to follow the obvious step of ending beef and dairy consumption or air travel? The most obvious step, and historically favored by political elites over the past 200 years, is to reduce the number of people on earth. So far, such utopian leaders have removed about 200,000,000 people from earth through global and civil wars, intentional starvation, and horrifying death camps. Do we really want more of that? Why couldn’t our authors bring themselves to consider or even acknowledge these sad possibilities? (END)