In Praise of Scientific Theory

Just a hunch? Hardly. Think germ theory, atomic theory and the theory of evolution.

Start with an observation.

Source: Hulton Archives/Getty Images

Science can make life difficult for manipulators and demagogues. Without science, it would be much easier to convince the public that an intelligent designer created the world, or that greenhouse gas warming and lead contamination are just the fantasies of “alarmists.” To physicist and historian Gerald Holton, attacks on science tend to go along with moves toward authoritarian rule. “History has shown repeatedly that a disaffection with science and its view of the world can turn into a rage that links up with far more sinister movements,” he wrote in his 1993 book “Science and Anti-Science.”

Those who want to fight the conclusions of scientific research often strike at its points of vulnerability -- like scientists’ insistence on using the word “theory” to describe even well-established ideas. In popular language, a “theory” implies a hunch or guess –- something less than a fact. That wrongly suggests weakness. “The theory of global warming is just that: a theory,” then-congressman and climate skeptic Mike Pence told an Indiana newspaper in 2003. He probably couldn’t get away with a similar dismissal of germ theory, atomic theory or Einstein’s theory of relativity.

“It’s unfortunate the way the word ‘theory’ is used,” said philosopher of science Peter Godfrey-Smith. “To say something is a theory is to say it’s been expressed as an idea. It’s not to say anything about whether the claim is justified or not justified -- true or false.”

The 19th-century philosopher William Whewell was one of the first people to put forward a definition of scientific theory as a product of observation and reasoning. And he had in mind a particular form of reasoning -- inductive reasoning, by which people draw broad conclusions from individual examples. That lines up with Darwin’s theory of evolution, which is backed by the fossil record, DNA evidence and even changes in animal anatomy.

For a theory to be scientific, the philosopher Karl Popper asserted, it must be testable in such a way that it could be proven false. “Creation science” and other ideas involving supernatural entities can’t be falsified by test (How do you prove God doesnt exist?), and therefore aren’t scientific Popper also argued that proper experiments can only be set up to falsify theories, not to confirm them. (Philosophers of science are still debating the merits and flaws of Popper’s ideas.)

Different fields have developed different standards for describing as-yet-untested ideas. The late physicist Richard Feynman called them “guesses.” In one of his famous lectures at Cornell University in 1964, he echoed Popper in declaring that guesses are legitimate parts of science if they’re falsifiable: “It doesn’t make any difference how beautiful your guess is; it doesn’t make any difference how smart you are,” he said. “If it disagrees with experiment, it’s wrong.”

This is true -- assuming the experimenters didn’t make a mistake, as sometimes happens. For example, a few years ago, a group of physicists found something that seemed to directly contradict Einstein’s theory of relativity, which predicts that nothing can move faster than the speed of light. In an experiment, it appeared that particles called neutrinos traveled faster. Other physicists were rightly reluctant to believe these particles were violating this universal speed limit -- not because Einstein was so smart or because the theory was beautiful, but because the theory itself is backed by dozens of experiments. In other words, they weren’t really weighing an observation against a theory; they were weighing a single, new observation against dozens of well-established observations. (And, in fact, the experiment turned out to be in error due to a faulty cable.)

The theory behind climate change is also grounded in observation and reason. It all started with a puzzle: By the early 1800s, physicists realized that an earth-sized rock orbiting the sun at a distance of 93 million miles should be frozen according to the known laws of physics. French physicist Joseph Fourier proposed that the atmosphere keeps the planet warm. Others tested this theory in laboratory experiments, sending a simulated version of sunlight through various gases. They found that oxygen and nitrogen had no effect on the light, but carbon dioxide did. In repeated experiments, carbon dioxide absorbed and re-radiated infrared waves, which on a planetary scale would prevent some of the sun’s energy from escaping to space.

But now climate science has something even stronger on its side, said atmospheric physicist Lee Harrison of the State University of New York, Albany. The premise is all predicted by a powerful theory in physics known as quantum mechanics, which describes in detail the behavior of light and matter on the scale of molecules, atoms and subatomic particles. Like Einstein’s theory, quantum mechanics is bolstered by hundreds of experiments. Quantum mechanics predicts how infrared radiation coming up from the earth will be affected by carbon dioxide and other gases.

“What the public doesn’t understand is the extreme interconnectedness of physical reality,” Harrison said. “If someone proposes that carbon dioxide is not a greenhouse gas, this requires ripping up essentially all of modern physics … Now you aren’t just arguing with those measurements of carbon dioxide; you are arguing with the whole body of molecular quantum mechanics and all confirming measurements.”  

The most insidious misperception skewing today’s debate over climate change is confusion between uncertainty about the predictions of a theory and uncertainty about the theory itself. To illustrate the difference, Harvard University philosopher Peter Galison brings up evolution by natural selection: The theory is on solid ground, but that doesn’t mean it can predict exactly what foxes will look like 800,000 years in the future.

New York Times columnist Bret Stephens conflated these kinds of uncertainty in his high-profile debut column last April: “We live in a world in which data convey authority. But authority has a way of descending to certitude, and certitude begets hubris,” he wrote. But science isn’t just data-collecting, which is why scientists can be close to certain of the merits of germ theory, relativity, quantum mechanics and evolution. Scientists are the ones avoiding hubris by continuing to call their best ideas theories.

In his column, Stephens wrote that “[ordinary citizens] know -- as all environmentalists should -- that history is littered with the human wreckage of scientific errors married to political power.” Holton’s “Science and Anti-Science” is full of examples of this -- from the Soviet anti-scientific biological theory known as Lysenkoism to the Nazis’ replacement of powerful 20th-century physics with ad-hoc cosmology. But there’s a big difference between these made-up theories and the experimentally backed theory of greenhouse gas warming. I’ll leave the last word to Holton:

History has shown again and again that authority which refuses belief in sound science and substitutes for it non-science in its programs can have vast costs in lives and treasures; think of the mass starvation when the Soviet party adopted Lysenko’s ideas for harvests ... or what might be coming now for the lives of coastal peoples if no action is taken in accord with sound climate science.

Read more in Faye Flam’s ongoing series about the science of sorting fact from falsehood:

Science’s Weakness Is Also Its Strength
Why Scientific Consensus Is Worth Taking Seriously
How Science Sorts Fact From Alternative Fact

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Faye Flam at

    Before it's here, it's on the Bloomberg Terminal.