How Does Science Work?
Today, I’m going to take a break from the coronavirus to talk about something broader. Why we can trust science. I believe a lot of modern misconceptions about science in general and certain scientific theories in particular stem from misunderstandings about how the scientific process actually works. This post will get a bit more abstract and generalized, but I believe understanding this, far more than any individual scientific theory or topic, is the most important to understanding how and why scientists do what they do and what scientific knowledge actually means.
Law, Theory, and Hypothesis
The process of science includes observing and describing phenomena, explaining the workings of that phenomena and conducting experiments of different aspects of our current understanding of a phenomenon in an effort to gain greater knowledge and push our understanding further. Let’s dive in and better understand the differences between a law, a theory and a hypothesis-key factors in the scientific process.
A common line from creationists is that evolution “is just a theory.” This is true, but it deliberately misunderstands what the word theory actually means, at least to a scientist. So let’s talk about the process of forming a theory.
In science, a law is a description (usually mathematical) of an observed phenomenon. For example, the Law of Gravity is a mathematical description of how two objects with mass will be attracted toward each other by the force of gravity. Issac Newton was the first person to describe gravity mathematically, calculating this equation:
He discovered this by observing with a telescope the motion of planets around the sun and moons around those planets and then measuring how fast they orbited compared to how close they were to what they were orbiting. He compared these measurements to each other and determined that they all fit this equation. The equations that Newton deciphered are still used today when charting space flights and launching satellites launches, so they’re very useful at describing what gravity does. But the Law of Gravity does not tell us what gravity actually is or how it works. That’s where a theory comes in.
So where a Law describes a phenomenon, a theory provides an explanation of the phenomenon based on existing data. This does not mean a theory is unproven! A theory can be either unproven, disproven, or accepted fact depending on how much evidence it has, but all theories try to describe the universe given what we know about it.
Albert Einstein was not the first person to propose an explanation for how gravity worked, but his Theory of General Relativity was the strongest. It wasn’t originally intended to explain how gravity worked, but by following the evidence, he eventually concluded that any object with mass will warp and curve the space around it. Now, in non-Euclidean geometry, lines that would otherwise never intersect with each other will intersect if they’re drawn on a curved surface. (Think of lines of longitude on a globe; they’re parallel to each other at the equator but they all intersect at the poles because they’re on a curved surface.) So if an object were to travel through the curved region of space around a second object (say a planet), its straight path would be curved toward it until the two paths met. Since we can’t see the distorted space, it just looks like the object is pulled toward the planet by an invisible force. (This video explains this well, but it is fairly complicated and by no means required reading.) So while Newton’s Laws explained how gravity behaves, Einstein’s theory explained why gravity behaves that way.
A hypothesis is the proposed explanation for a very specific observation or situation made with the intention of being proven via experiment. When a scientist makes observations of a particular phenomenon, they try to come up with explanations for why this phenomenon occurs as it does and how it works. A good hypothesis is a simple “if-then” statement; if X is true, then Y will happen when I do Z. An example might be “if I increase the amount of light a houseplant receives, then it will grow larger.” The next step would be to design an experiment that allows one to measure this phenomenon as it occurs. For example, you could take two of the same type of houseplants and place one in a sunny windowsill and one in a shady corner, and measure their heights over the course of a few weeks. If the hypothesis is true, the sunny plant will have grown more than the shady one in the same amount of time and there will be quantitative evidence that this occurred. A strong theory is one built off of numerous proven hypotheses to create a useful model of how some phenomenon works.
When Einstein first published his theory of general relativity in 1915, it was met with skepticism. While all the math of his proposal made sense and fit with known physics, the notion of space and time being warpable was a revolutionary departure from some very old assumptions about the universe. Newton pictured space as a static void, quite unlike general relativity’s flexible fabric-like vision of space-time. For one, Einstein’s theory being true would mean that anything moving past a massive object, even something massless such as light, would have its path changed by traveling through this curved space. But in order to prove this, scientists would have to observe light traveling past an extremely massive celestial body that was also close enough to Earth for the distortion to be observed. A celestial body such as the Sun.
Astronomers Arthur Eddington and Frank Watson Dyson started working on an experiment in 1917 to prove Einstein’s theory. Their hypothesis was that if mass did curve space-time, then the Sun’s mass would distort the path of light from stars behind it. If a star was close enough to the Sun in the sky, then it would appear to be in a slightly different place than at night because the star’s light was traveling through the curved space around the sun. This couldn’t be easily tested because...well, you can’t see stars during the day. So they waited until the total solar eclipse of 29 May 1919 and sent two teams to Brazil and West Africa to photograph the total eclipse. Sure enough, the stars around the edge of the Sun were in slightly different positions than they were normally in the night sky, proving Eddington and Dyson’s hypothesis. While the Theory of General Relativity already had been shown to be mathematically sound, this was the first experiment to support it with hard observable evidence.
Einstein himself was not present for these tests, but this confirmation of general relativity made him world famous. When later asked what he would’ve done if the experiment hadn’t proved him right, he was quoted as saying, “Then I would feel sorry for the dear Lord. The theory is correct anyway.”
Strength of a Theory
While the Eddington Experiment was the first experiment to support General Relativity, it would not be the last. Even before this experiment, relativity was able to explain several phenomena that Newton’s laws couldn’t. There were other theories of how gravity works before Einstein, such as invisible particles knocking objects closer to each other or whirlpools in an invisible interstellar aether, and all of these fit Newton’s Laws well enough, but none were ever able to be proven experimentally. Nowadays, General Relativity is the only theory of gravitation with any real support among mainstream scientists because it meets so many of the criteria for a strong theory. There are several ways to gauge the strength of a scientific theory.
What Else Does It Explain
Does the theory explain a broad diversity of phenomena? I find a good way to think about this is to ask, “For this to be wrong, what else would have to be wrong?” The more things that would also have to be wrong, the less likely the theory is wrong. For example, Darwin’s Theory of Evolution by Natural Selection is known for having a history of non-scientific contention on ideological grounds. But among the scientific community, it is one of the strongest theories there is because of the breadth of what it explains. Basically, everything we know in all of the life sciences, in biology, genetics, ecology, paleontology, and much of what we know in psychology, sociology, and medicine, only makes sense from the perspective of evolution by natural selection. This isn’t necessarily to say that evolution can never be disproven, it’s to say that every fact we know about all of these fields of study counts as a point in evolution’s favor. If someone wanted to disprove this theory, it wouldn’t be enough to poke holes in it; they would have to create a better theory which explains everything evolution explains but better. This isn’t impossible, but it would be a feat.
Simplicity of Assumptions
Another gauge for the strength of a theory is its simplicity. This does not mean that the best theories are the easiest to explain, but rather the best theories follow Occam’s Razor. Occam’s Razor is the principle that if one has two explanations for a phenomenon and they have an equal amount of evidence, the explanation that makes the fewest assumptions is more likely to be correct. Let’s take the moon landing hoax as an example (though it isn’t a scientific theory). If we assume that there is equal evidence that the Apollo moon missions were fabricated as they were genuine (there is plenty of evidence they were real), there are many other things which would have to be true for such a hoax to be committed. Such as;
NASA would have to have built and launched all the Apollo rockets in order to keep up appearances, likely including test rockets and missions. Even if these rockets weren’t carrying astronauts, this would’ve been similarly as expensive as a manned launch.
All the film and photographs taken on the lunar surface would have to been doctored, either with anachronistic lighting technology or airbrushing that can’t be detected 50 years later. (This clip from the show Adam Ruins Everything explains this well.)
Hundreds of thousands of people would need to be sworn to secrecy, from NASA employees to members of foreign governments, with no leaks whatsoever in over half a century.
Compare this to the assumptions we would have to make if the landings were real;
The wealthiest government on Earth managed to build a tube full of explosives, strapped a few trained pilots to the top, and sent them on a wildly dangerous mission for the sake of international prestige and exploration.
Again, evidence could be disclosed tomorrow proving any of these assumptions were true, which could make the hoax theory stronger. But assuming the Moon landings were a hoax with the evidence we have today isn’t sensible because there’s so much else you’d have to believe without evidence. And I’m sure there are many other conspiracy theories that you can think of which would fall apart when you consider Occam’s razor.
Prevalence of Support
Lastly, a good gauge, albeit an indirect one, for the strength of a theory is what percentage of the relevant scientific field buys into the theory. The more support a theory has from the relevant scientists, the more trust you can have in it. Presently, 97% of climatologists believe that man-made climate change exists and is a genuine threat to our society. There are climatologists who don’t believe climate change is a threat, with 7.8 billion humans out there, you can find every possible opinion represented somewhere. But as we’ve discussed before with Andrew Wakefield, one person in a lab coat saying something does not make it true. A scientist can be biased, contrarian, or just plain mistaken, just like any other human, but that is why we have the scientific method and the scientific institutions that we have. Scientists are expected to carefully plan out their experiments according to strict guidelines, detail every step of the experimental process, and publish all of these writings in scientific journals available to other scientists. This allows other scientists to review their findings, find flaws in their experiments, and even perform the experiments to see the results for themselves. This helps to filter out a scientist’s personal biases and to naturally select for ideas that are thought out and objective. Now, this process is far from perfect; it is very hard to get funding for replication studies (when a scientist does someone else’s experiment to confirm its findings) because they aren’t as sexy as original research. And as the history of eugenics shows, it’s much harder to filter out personal bias when only one group of people are allowed to be scientists. But these are problems with the infrastructure of science that scientists are addressing because it will improve our ability to gain knowledge and understand what is true. By working together and verifying each other’s work, we gain a better understanding of the world around us and can use that knowledge to make the world a better place.
Comments