The Day Science Was Overthrown

by Brian Dunning
October 18, 2022

The scientific method is basically the process of having a question, finding an educated guess that seems to answer it, then testing that answer and — 99% of the time — finding it to be wrong. But eventually we find an answer that withstands all our attacks to find its errors. So we consider that to be our provisionally correct answer, at least until a better one comes along. Some fields of science are so well established, and so thoroughly proven by experiment, that no better answers are likely. But sometimes — once in a long while — we get surprised. Today's episode is a look at some of the best such surprises.

Unfortunately a lot of people are hostile to science, often because they have some pet belief that is not supported by science — so they blame the process by which we learn, rather than allowing for the possibility of their own error. Often they will assert that scientists are motivated by the need to maintain some status quo, to refuse to consider anything new. The justification for this is often the claim that their grant money would be threatened if they explored any new directions: as if grantors invest money in the hope of learning nothing!

Imagine the working research scientist, who on his first day on the job as a postdoc, states his goals: "I will find success by learning nothing new, by making no contributions to my field; but by reverently abiding by all my predecessors' work and maintaining the dogma of the status quo." Do you know any researchers whose professional aspiration is to have their whole career produce nothing at all? Of course not, because human beings are the opposite of this. For a research scientist to find a flaw in accepted work and to make a significant change in what we know can be the apex accomplishment of a career. Breaking out new paradigms is the dream of every research scientist. Very few will win that lottery, but all will spend their careers playing it. One of the reasons the scientific method works so well is that this human self-interest and desire to succeed intersects beautifully with the need for continuous self-scrutiny and self-correction that is crucial to the evolution of any sound science.

The conspiracy theories of those hostile to science could not be more wrong, and the constant growth of our body of knowledge is the proof.

So today we're going to look at examples of this in action: cases where sound science had been accepted, and then new data and new discoveries upended work that had been considered solid yesterday, and thus improved our knowledge of the world. Some look at this constant improvement as proof that the very scientific method itself is hopelessly flawed, with the view that if science was wrong yesterday then it's wrong today, therefore anything coming from the scientific method is by definition untrustworthy, and that we should turn instead to intuition, to the metaphysical, to the spiritual. Well, that perspective is half right. They are right in observing that much of what we know now is likely to be improved; but they are wrong in believing that it's best to leave that path and go instead in the opposite direction.

What follows is a list of my favorites, and like all lists, it's incomplete; so don't email and tell me I left something out. Lots of other examples are left out due to time constraints, if nothing else. But also I'm leaving out examples that don't fit the bill. We're looking at cases where theory was based on solid science, which then got turned on its head. We're not including cases like Galileo where what got turned on its head was church doctrine, not solid science. We're not including cases like Pons & Fleischmann's cold fusion, which never had solid science behind it, only bad science. And we're not including things like the four bodily humors, which was based on pre-scientific ignorance, not on science.

1. Peptic Ulcers

The classic case, and one of the best known, is the famous change in our understanding of what causes peptic ulcers. It had always been understood to be caused by environmental factors like spicy food, stomach acid, and particularly stress. But in 2005, researchers Barry Marshall and Robin Warren were awarded the Nobel Prize in Physiology or Medicine for their discovery of the bacterium Helicobacter pylori and finding that it is the primary cause of such ulcers. Marshall actually swallowed a culture of it himself just to make sure. He was correct. Like many of these changes in science, it was not widely accepted overnight, but took time for the gastroenterology community to embrace. But treating the bacteria worked and treating the stress didn't, and so the consensus ultimately did shift — just the way science is supposed to work.

2. What Killed the Dinosaurs

Today, the generally accepted cause for the extinction of the dinosaurs, called the K-Pg extinction, is the impact of the Chicxulub asteroid on the coast of Yucatán about 66 million years ago. This is a fairly recent discovery; the telltale layer of sediment (called the K-Pg boundary layer) was found by Luis and Walter Alvarez only in 1980. It was 30 years later, in 2010, that an international panel of scientists finally signed off on the K-Pg impact event as the primary cause. Previously a multifactorial cause was the leading theory, and which is still held by some today. These factors, all of which are solid and well accepted individually, include volcanism adding particulates to the atmosphere, sea level recession causing widespread climate change, and the formation of flood basalts in western India releasing sulfur dioxide.

3. Plate Tectonics

Today we know that the Earth's lithosphere consists of a number of large continental plates which grind past each other, and even climb over and under each other, as the mantle dynamics beneath them move them around, and forces of gravity and the Earth's rotation also contribute. But it was not always so. Until the movements at the oceanic ridges were confirmed in the 1960s, the prevailing theory was called continental drift, which posited that the movement of the land masses through the crust was more like icebreakers going through ice, but without any agreed mechanism. And even this was a relatively new theory, having been developed in the early 1900s by German geophysicist Alfred Wegener. Before that, scientists were staring at the interlocking fit of the coastlines on globes, scratching their chins. For about 300 years, geographers figured that the continents moved around, but with no notion at all of how often, on what paths, or what drove them.

4. Taxonomies of Organisms

Carl Linnaeus was a Swedish multi-disciplinarian who, in 1735, published the taxonomic system he had devised to organize our understanding of natural organisms. He began with three kingdoms: animal, mineral, and vegetable; and divided them into five ranks: Class, Order, Genus, Species, and Variety; later refined by other naturalists into Kingdom, Phylum, Class, Order, Family, Genus, and Species. This worked well for centuries (and in fact is still largely used today), but a major paradigm shift took place with the discovery of genetics and the resulting modern evolutionary synthesis. Every year we find more and more cases where an organism can no longer neatly fit into Linnaean classification. Today nearly every biologist recognizes that ranks have no real utility. So although we thank Linnaeus for what he did at the time and for how much help it was then, we no longer need it. The system of using a two-part name for every species was also devised by Linnaeus, and we do still use it today.

Correction: In the recording I misspoke and read "Swiss" where "Swedish" was written. No excuse other than a brain fart. The recording has been updated. Apologies to the Swedish listeners. —BD

5. Einstein's Cosmological Constant

The history of cosmology is packed with examples that could fill ten episodes like this. But we'll focus on just one today. In 1917, the accepted cosmological model was called the static universe, in which space is infinite, time is infinite, the universe is neither expanding nor contracting, and it's flat without any spatial curvature. It's basically what a universe would look like without any weird stuff, like what a reasonable person would create if they could be God for a day. Albert Einstein was one of its adherents, and in one of his equations that modeled the universe, he added a number called the cosmological constant as a fudge factor to cancel out the way that gravity would otherwise draw a static universe together on itself. Then in the 1920s, astronomer Edwin Hubble proved the universe was expanding, much to his own disbelief; and Einstein realized his cosmological constant was unnecessary — later calling it the biggest blunder of his career. And then, in the 1990s, it was proven that the expansion is accelerating, and cosmologists are looking again to Einstein's cosmological constant to give it a positive value — which we're equating with the placeholder term "dark energy". Are we going to have to turn everything on its head again tomorrow? I don't know, but the chances are non-zero.

6. Dinosaurs Were Cold Blooded

Most reptiles are cold blooded (ectothermic in technical terms), meaning their body temperature relies on environmental sources of heat; and so, for most of the history of our knowledge of dinosaurs, it's been accepted as a given that they were cold blooded too. Being cold blooded is really efficient; you don't use up energy running an internal heater. Being warm blooded, or endothermic, means you use additional energy to crank up the metabolism when needed to warm up. But these days, we have newer, more detailed knowledge. We now know that birds — which are descended from dinosaurs — are warm blooded; we also know that warm or cold blooded is a false dichotomy. There are many different ways that different species regulate their body temperature. Some are homeothermic, poikilothermic, mesothermic, heterothermic, stenothermic, eurythermic — but I'll just stop here. It's rarely perfectly accurate to describe most given species as being purely of one type or another. But substantial research, mostly in just the past few years, has now taught us that most dinosaurs probably weren't strictly ectothermic. Among this is study of dinosaur eggshell fossils subjected to a test called clumped isotope paleothermometry. Long story short, you can determine the mother's internal body temperature by looking at the order of oxygen and carbon atoms in a fossil eggshell. Is that wild? Yes. Were all dinosaurs cold blooded? No.

That's enough for today. If you've got any other favorite examples — respective of the criteria we laid out up top — send them my way, at brian@skeptoid.com. Maybe we can put together a part 2. But regardless, take these examples as nothing more or less than what they are: real world examples of the scientific method in action. Everything you know now that is the result of the scientific method changed uncounted times before you learned it; and it will continue changing long after you've forgotten it. That's why it's best: it changes, at the drop of a hat, the instant new data comes in and is confirmed by the community. This is a challenging concept for many, but if you can keep your mind open to whatever we can learn by following the scientific method, you'll be right far more often than you're wrong. Embrace what we can learn by the best method we know of, and live in a state of perpetual excitement that something you know — and none of us can guess what it might be — will change tomorrow and explode into something newer, smarter, and better. That is the excitement, and the passion, of science.

By Brian Dunning
Follow @BrianDunning