Galileo Gambit

27 Galileo Gambit

David Kyle Johnson

Almost everyone thought Galileo was wrong, but he turned out to be right.

Darin Brown (HIV skeptic) The Galileo gambit is an argument sometimes made in defense of theories that contradict the mainstream scientific consensus. Disagreeing with a consensus is not itself the fallacy. Such theories are often ridiculed or even laughed at. But to those who commit this fallacy, this is actually a reason to think that their theory is true. “After all,” they reason, “the mainstream laughed at Galileo when he said the sun was the center of the solar system; that flew against conventional wisdom too, but that turned out to be right. So my theory is probably right too.” The fact that their fringe theory is ridiculed is thought to be a good reason to conclude that it is true.

The name also simplifies Galileo’s history, which involved both religious authority and disputes about astronomy. But there are examples of those who did contradict scientific consensus who turned out to be right. Louis Pasteur and his germ theory of disease, Alfred Wegener and his theory of continental drift, and Albert Einstein and his theory of relativity are three prime examples. But since Galileo is the scientist that those who commit this fallacy invoke, that is what it is called.

Regardless, there are many mistakes that underlie this line of reasoning. The one that perhaps most obviously exposes the fact that it is fallacious is related to confirmation bias (see Chapter 73), availability error (see Chapter 21), and suppressed evidence (see Chapter 98). The person committing the fallacy is drawing attention to one case where the dissenting view turned out to be right but ignoring the many cases where the dissenting view turned out to be wrong.

It’s not surprising that such cases are easily ignored; they are very hard to find because we don’t usually learn about the theories that turned out to be mistaken. A comparison must also include dissenting claims that failed, such as those of Urbain Jean Joseph Le Verrier (who proposed a real planet Vulcan), Wilhelm Reich (who essentially proposed that libido was an actual “physical energy” released upon orgasm), René Blondlot (who proposed the existence of N‐rays), and Martin Fleischmann and Stanley Pons (who said they had achieved cold fusion; see Schick and Vaughn 2014; Shermer 2002). As Carl Sagan (1984) put it, “The fact that some geniuses were laughed at does not imply that all who are laughed at are geniuses. They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown” (64).

Ridicule alone does not tell us which kind of case we are dealing with.

Perhaps the best way to describe the fallacy is as an association fallacy or a faulty analogy. The mere fact that two things have one property in common doesn’t necessarily mean that those things have a second property in common. Now it does if these properties are related to each other. Observing that, like humans, rats are mammals with lungs justifies the conclusion that rats breathe like humans do. That’s a relevant similarity. The fact that both humans and rats have lungs doesn’t, however, entail they can both do calculus. And that’s what the Galileo gambit does: it relies on an irrelevant similarity or association. Yes, Galileo was criticized, but that has nothing to do with why his theory was right. So the fact that your theory is criticized tells us nothing about whether or not it is right.

The content of the criticism can tell us something. If a proposal conflicts with a well-supported scientific consensus, the existing evidence gives us reason for skepticism. Whether it be the Wright brothers (and their flying machine), Wegener (and his theory of continental drift), or Einstein (and his theory of relativity), even though they turned out to be right, their contemporaries were right to be initially skeptical. Coherence with established scientific belief is an important criterion by which scientists can and should judge proposed theories. It’s called conservatism. It’s not the only one, which is good because otherwise we could never learn that we are wrong about something and advance our understanding of the world. When a theory conflicts with established evidence, it needs to explain that evidence and offer grounds for preferring it. Withholding acceptance leaves room to investigate the proposal.

How could such a theory prove itself to be the better explanation? According to Ted Schick and Lewis Vaughn (2014), in addition to conservatism, science uses four criteria to weigh competing theories:

Testability: Does the theory make observable novel predictions?

Fruitfulness: Does the theory get those predictions right?

Scope: Does the theory have explanatory power? Does it unify or expand our knowledge? Or does it just invoke the inexplicable or raise unanswerable questions? The more a theory explains, the wider its scope.

Simplicity/Parsimony: Does it make additional assumptions? Does it require the existence of forces, energies, or things that we don’t already know exist? The fewer such assumptions, the simpler a theory is.

These criteria guide comparisons; they are not a simple tally in which satisfying the largest number automatically settles the issue. So even if a theory is not conservative (that is, it conflicts with the consensus view), if it proves over time to be more fruitful, wider scoping, and more parsimonious, then one should accept it.

This is what happened with Einstein’s theory of relativity. At first, it was not conservative because it conflicted with Isaac Newton’s theory of gravity, which was very well established. Relativity accounted for the bending of starlight around the Sun and the unexplained part of Mercury’s orbital precession more successfully than the Newtonian account. These made relativity more fruitful and wider scoping than Newton’s theory. Later we confirmed time dilation and black holes. In 2016, the first direct detection of gravitational waves was announced, providing another test of general relativity. Of course it took time; I’m sure some stubborn old school Newtonian scientists never admitted that Newton was wrong, but eventually the scientific consensus changed. Now if a theory challenges relativity, scientists are rightly skeptical of it.

That of course is not to say that Einstein’s theory is perfect. There will likely be an improvement upon it, just as Einstein’s was an improvement on that of Newton. But Einstein likely will never be shown to have been completely wrong and misguided. Generally this is how science works. A successor theory often retains an older theory’s successful predictions within a limited domain. The important demand is that a proposed replacement account for the evidence the older theory explained, as well as any evidence it did not.

The author gives several examples of the Galileo gambit. It’s very common in the alternative medicine community, especially among those who deny the efficacy of vaccines and claim that they cause conditions such as autism. Vaccination has a substantial evidence base, so a claim that contradicts it needs evidence (see CDC 2015). A specific claim about harm should be evaluated on its evidence, rather than dismissed or accepted because it challenges a consensus. For example, if the rate of autism was higher in vaccinated vs. non‐vaccinated kids, we’d want to investigate further. But it’s not. Indeed, countless studies have been conducted, all showing no relationship between vaccines and autism (see CDC 2015).

Yet anti‐vaxers won’t back down. They even gave an award to Andrew Wakefield, the author of the fraudulent study that started the whole “vaccines cause autism” scare. And you’ll never guess what they called it. “The Galileo Award.”

Other examples the author identifies include Rupert Sheldrake (2004) and his morphic resonance/fields, Bigfoot believers, UFO enthusiasts, creationists, conspiracy theorists of all stripes, and (most dangerously) people who deny climate change.

Darin Brown, the HIV denier quoted at the start, makes a somewhat different argument. What follows from the example of Galileo, he thinks, is not that his position (that HIV doesn’t cause AIDS) is right, but that “just because almost everyone thinks [you are wrong], doesn’t make it so.” He claims that those who argue he’s wrong by quoting the scientific consensus (The Institute of Medicine, The World Health Organization, the CDC, … every major scientist and relevant expert) are committing an “argument from consensus fallacy.” Invoking Galileo is just his way of showing the error of their ways – that the scientific consensus does not “make it so.”

But citing relevant scientific experts as evidence for a scientific claim is not fallacious. Neither is invoking a well-supported scientific consensus to cast doubt on a new theory. Sure, not even scientific consensus could ever prove anything 100%, but hardly anything ever can and no one was claiming it did. A consensus grounded in converging evidence can provide strong reasons to accept a claim. Its force comes from that evidence and relevant expertise, rather than agreement alone.

Brown is actually confusing (what he calls) the “consensus fallacy” with the ad populum fallacy. The mere fact that a large group of people believe something is true is not a good reason to believe it; that’s right. Historically, large groups have been wrong about a lot of things. But if that large group of people is a group of experts on the topic at hand with mounds of evidence for their position, that is a good reason to accept their claim and to be dubious of claims that contradict it. A consensus remains open to revision; how much confidence it deserves depends on the evidence supporting it. When you are contradicting the consensus of experts who are better educated and much more versed than you are in the topic at hand, it’s much more likely that you are the one that is wrong.

The Galileo gambit is also used to suggest that science is closed to criticism. Scientific methods make criticism and testing central to evaluating claims, even though individual scientists can resist them. But it’s also not fair to demand that scientists consider every contrary theory out there; there are just too many. Scientists would never get anything else done. As Michael Shermer (2002) has noted:

For every Galileo shown the instruments of torture for advocating scientific truth, there are a thousand (or ten thousand) unknowns whose “truths” never pass scientific muster with other scientists. The scientific community cannot be expected to test every fantastic claim that comes along, especially when so many are logically inconsistent. (50) So if you want your theory to be taken seriously, the burden is on you to do the initial work of proving that it is better than the established view. The mere fact that your theory is contrarian is not enough; indeed, that means it’s likely false. If a scientist won’t consider your theory before you have shown it worthy, that refusal can reflect a reasonable allocation of limited time. It does not, by itself, establish the theory’s truth or falsity. As Robert L. Park once put it, apparently as an offhand remark in an academic bulletin for the University of Maryland, “To wear the mantle of Galileo, it is not enough that you be persecuted by an unkind establishment; you must also be right” (see Shermer 2002, 50).