Why do scientists believe in untestable ideas?

And what do we do about it?

Why do some scientists believe in untestable ideas? That’s the question being asked by philosophers of science.

Physicists have long relied on a notion advanced by philosopher Karl Popper, that a theory is scientifically valid if it is falsifiable. But in recent years, many serious physicists seem to have abandoned this model. String theory, for example, is one of the most exciting ideas in modern physics. But it’s not testable—so how can physicists be confident that it’s sound?

String theory is exciting in that it provides a way to reconcile two seemingly incompatible but equally successful ideas in physics—general relativity and quantum mechanics. The problem is that string theory is currently physically untestable, which means no matter how well its own internal logic may work, we don’t know if it actually models reality. That makes the name “string theory” ironic, since it’s more properly understood to be a hypothesis.

There are two general reasons a hypothesis might be untestable. One reason is practical: the technology to make the necessary observations doesn’t yet exist. Think of the ancient Greeks who didn’t have telescopes and therefore couldn’t test the idea of stellar parallax. The other reason is by definition: no conceivable progress in technology will ever allow the necessary observations or experiments. The multiverse falls into this category. How do you test something that’s outside of and causally detached from this universe? By definition, you can’t.

So, which category does string theory fall into?

String theory is untestable for a number of reasons. It contains an infinite array of possibilities with no way to distinguish between them, it requires a new geometry that we haven’t figured out, and its predictions are both imprecise and at a scale that no experiment is capable of achieving. We currently have no idea of any equipment or experimental methods, even in principle, that would allow us to deal with all of that. But does this mean they don’t exist? I don’t know. But any hope, for now, is extremely thin.

With so little hope of advancing string theory from hypothesis to theory, why has it persisted as a subject for decades? And why, for that matter, has the multiverse also persisted for so long? They seem to be part of a movement in the physical sciences to accept untestable ideas under the umbrella of true science, something the old guard is not happy about. But, contrary to popular sentiment, the reasons for this movement are not arbitrary.

According to philosophy of science researcher, Richard Dawid, there are three reasons a physicist will believe in an untestable theory:

1. The theory is the only game in town; there are no other viable theories.

2. The theoretical research program has produced successes in the past.

3. The theory turns out to have even more explanatory power than originally thought.

Any of these arguments by themselves is not enough to convince a physicist that an untested theory has merit, but all three together are pretty powerful. That said, this powerful combination doesn’t replace empiricism as the gold standard for determining scientific truth.

A lot of the low-information “I love science” nerds believe that science is a continuously upward-moving entity, an enterprise perpetually rocketing in the direction of progress. But I believe we’re in serious danger of circling back to the proto-scientific methodology of the ancient Greeks, who, out of mistrust for experience, relied on thought experiments and famously got many of the ideas we now take for granted in physics completely wrong. While it’s true that our perceptions can be subjective, the history of science clearly points to the superiority of thought + empiricism over thought alone in understanding the workings of the universe. After all, it wasn’t just Einstein’s beautifully mathematical ideas that made him a scientific rockstar, it was the experiments that agreed with his ideas to high precision.

As scientific standards seem to erode, what can be done to ensure experimentation remains the gold standard in the physical sciences? We first have to identify the reason empirical support in science seems to matter less and less. I believe it’s because the empirical nature of physical science is rooted in the set of ideals and assumptions unique to Christianity. As science and our society are increasingly divorced from their Christian roots, that empirical nature not only becomes less important, but even becomes an obstacle to pet hypotheses and cherished beliefs. To paraphrase physicist Isidor Isaac Rabi, faith in the hard, crude methods of physics have always been a buffer against self-delusion.

I’ve written extensively on the subject here, but in short, modern science was born of Christian ideals and assumptions, and therefore almost certainly requires that foundation in order to persist and thrive. For that reason, I strongly believe Christians need to stop retreating from science. Instead, as a body, we need to become scientifically literate and engage modern science in a meaningful way.

This is an edited and updated re-post of an article from my old blog.