How Science is Undone

Clifford Burdick (1894-1992) spent his life trying to disprove modern evolutionary and geological theory. He was a creationist. And one of the few creationists in the middle of the 20th century with some training in geology.

What if, for instance, evidence of human activity, such as preserved human footprints, was found in the same geological strata as dinosaur fossils? According to modern science, dinosaurs lived hundreds of millions of years before humans evolved. But geologic evidence of co-existence – as Burdick alleged that he had found – would puncture that modern evolutionary picture.

Or, what if he found pollen in geological strata aged long before the supposed arrival of flowering plants, as he did later on, while working with a professional geologist? This, too, could be the critical failure of modern scientific understandings.

Burdick pinned his hope not just on these unusual findings, but on a particular philosophy of science: Popperian falsificationism (most famously described in Karl Popper's The Logic of Scientific Discovery).

Falsificationism deserves a much fuller treatment than I'm giving it here, but the quick and dirty version is that what distinguishes science from other modes of inquiry or discovery is critical tests of the ideas in question. These tests do not test to see if the idea is true because, according to Popper's way of thinking, no amount of evidence can prove scientific theories true. There may always be some evidence out there, which we haven't seen yet, that would invalidate the theory. But we can logically "prove" theories false – because any (good) evidence that contradicts the theory "proves" the theory false. And scientific work proceeds by attempting to falsify theories. We have a lot of confidence in theories that scientists have, after many attempts, "failed to falsify".

According to Burdick's way of thinking, the two findings mentioned above would falsify modern geological and evolutionary theory. Modern theory says, "this can't happen." But, look! It happened. Ergo, we should throw away modern theory and favor a creationist account, which would account for such evidence.

The typical foil to falsificationism has been Kuhnian paradigm shifts (as first described by Thomas Kuhn in The Structure of Scientific Revolutions) or, if you want to be trendy, Ludwig Fleck's "thought-collectives" (described in Fleck's Genesis and Development of a Scientific Fact), which, like Kuhn did decades later, argued for a more social and historical understanding of how scientific inquiry proceeds.

But we don't have to even get there. There's a far more parsimonious reason that Burdick's aspirational application of falsificationism fails: young earth creationism is just not a very good explanation for the data that we see. And, while modern evolutionary and geological theory may not be perfect, they explain a lot of that geological and biological evidence.

Officially, this way of thinking about scientific thinking is called "inference to the best explanation" or IBE. As scientists – or even just everyday thinking people – we should go with the best explanation we got, all else being equal. In other words, the working explanation we favor at any particular moment doesn't have to even be a good one – it just has to be better than anything else we got.

What makes for a good explanation? I wish more people would ask themselves this question. There is no definitive formula. But the breadth of the explanation – the amount of stuff that it explains – is part of it. Explanations that explain more stuff are good. The parsimony of the explanation is another factor. We favor simpler or more grounded explanations over elaborate explanations that require us to posit extra mechanisms or extra concepts or extra interactions.

Putting aside, for the moment, that the "dinosaur and humans living together" finding and the "pollen before flowering plants" finding were a combination of Burdick's wishful thinking and his less-than-competent geology, these two findings would barely put a dent in modern evolutionary and geological theory. Would the findings be odd? Yes. Would they contradict the theory? Certainly. But modern theory would still be the best explanation of all the evidence. Because there is a LOT of other evidence out there that suggests the earth has been around for a long time. And that evolutionary processes have been occuring for several billion years.

All Burdick's findings would be – at least pending more evidence – are curiosities. Anomalous evidence. Anomalous evidence is ubiquitous in scientific inquiry of every sort. It might contradict generally accepted ideas, but not enough to seriously question those ideas – often because there are other parsimonious explanations that better explain that anomalous evidence. Like someone accidently contaminating a geological sample, as Burdick seemed to have done when he found pollen in strata earlier than the existence of flowering plants.

Just in case I haven't repeated myself enough: we don't favor explanations because they are perfect; we favor them because they're good enough. Good enough at explaining what's going on to our satisfaction; good enough at predicting what's going to happen next; good enough, when appropriate, for us to control aspects of what the theory is explaining.

Drastic changes in our understandings of the natural and social worlds are not undone by some random bits of evidence "falsifying" a theory. They occur because of widespread agreement that a different explanation is better at explaining all of the relevant evidence. Physicists did not throw out Newtonian physics because it couldn't predict mercury's orbit. They threw out Newtonian physics – or, rather, preserved it in the old people's home for physics theories – when relativity could explain all of the stuff that Newtonian physics could and more.

References

My understanding of the creationist movement comes pretty much entirely from:

Numbers, R. L. (2006). The creationists: from scientific creationism to intelligent design. Harvard University Press. You can find lots about Clifford Burdick there.

Below are the three philosophy of science books I mentioned. The years are all the updated editions. Fleck's was originally published in 1935, but not translated into English until much later; Popper's in German in 1934, and in English in 1959; and Kuhn's in 1962.

Popper, K. (2005). The logic of scientific discovery. Routledge.

Kuhn, T. S. (2012). The structure of scientific revolutions. University of Chicago press.

Fleck, L. (2012). Genesis and development of a scientific fact. University of Chicago Press.

Lots of philosophers have weighed in on the idea of inference to the best explanation. I find Peter Lipton's work to be thoughtful and well-written.

Lipton, P. (2017). Inference to the best explanation. A Companion to the Philosophy of Science, 184-193. Currently available here.