Rethinking the Scientific Method
“Scientific discovery and scientific knowledge have been achieved only by those who have gone in pursuit of it without any practical purpose whatsoever in view”
On a warm June night in 1878, Russian chemist Constantin Fahlberg sat down for his evening meal. Unbeknownst to Constantin, his dinner would change history forever. As he began to eat a bread roll, he noticed something unusual, it was sweet. This was confusing since he had never tasted bread that was so sweet before. How was this possible? It soon became clear, and it was only a matter of time before Constantin realized what was going on.
At the time, Constantin was working for the H.W Perot Import Firm in a laboratory run by Ira Remsen, a professor of chemistry at Johns Hopkins University. A German immigrant, Constantin had spent most of his life in Europe studying chemistry. He was fluent in many Indo-European languages and had a clear passion for science. However, the night of his fateful sweetbread-encounter would ensure him a place in future chemistry textbooks, a wikipedia page, and prevent him from being forgotten as just-another-chemist.
After he tasted the bread, he ran to his lab where he spent the next several hours examining everything around his work station – beakers, vials, etc. – until he discovered the source of the sweetness. It turned out to be “an overboiled beaker in which o-sulfobenzoic acid had reacted with phosphorus (V) chloride and ammonia, producing benzoic sulfinide.” Constantin had previously synthesized benzoic sulfinide, but he certainly did not think to taste it – why would he? But serendipity set in that night, and the first artificial sweetener was born: Saccharin.
You now know benzoic sulfinide from those little pink “Sweet N’ Low” sugar packets, which can be found at your local diner, Denny’s, or IHOP. While Constantin’s story is remarkable, it is not as uncommon as you think. Like many other scientific breakthroughs in history, Saccharin was discovered by accident. Take Penicillin, for example. It was stumbled upon by Alexander Fleming in 1928 when he found it growing mold on an old experiment as he was cleaning up his laboratory (perhaps more astonishing is the fact that it took Fleming over a decade to convince health officials of its potential). Viagra, the microwave, chocolate chip cookies, the list goes on and on. Some of the most beneficial discoveries have been mere happenstance.
If you have had any interest in science then perhaps you’ve heard this story before: scientists does y but gets x, and x turns out to be exceptionally useful. Histories accidental discoveries is a popular science topic and it has been the subject of several books. However, have you ever thought about the implications of all this serendipity for the scientific method?
Think back to your middle school days when you were taught that all scientific inquiry begins with a prediction, which is then tested in an experiment, later analyzed, and concluded. Makes sense, right? But when we consider that most discoveries weren’t in the initial plans, i.e., Saccharin, the scientific method seems to misunderstand how science works. A more accurate understanding realizes that randomness and chance are highly relevant steps in the scientific method. Put differently, within its logical and structured step-by-step procedure, the scientific method doesn’t leave room for the Saccharins and Viagra’s of history. This brings me to my first point: the fundamental problem with the scientific method is that it assumes you can predict the future. Let me explain.
Imagine that you are a cave man or woman living somewhere in Europe around 10,000 BC when your local chief tasks you with the responsibility of predicting the major scientific discoveries of the next several thousand years. Without a doubt the invention of the wheel should be one of the first things to make the list. But here is the catch: The ability to predict the invention of the wheel presupposes knowledge of the wheel. In other words, if you could predict the wheel then you might as well build it right then and there since you already what it is. As Nassim Taleb says in The Black Swan, “if you know about the discovery you are about to make in the future, then you have already made it.”
This is the fallacy of the scientific method: it doesn’t realize that implicit in a prediction about the future, albeit about the wheel or artificial sweetener, is knowledge of the future.
My other problem with the scientific method relates to how we think of the past. When we learn about the scientific method we get the idea that discoveries were made because it was followed. This is a mistake, and it brings me to my second point: our explanations of how scientists made their findings are warped by us knowing that they actually happened. These explanations are problematic because they are what psychologists call post hoc explanations.
Poc hoc explanations are descriptions of the past that are based on knowing what has already happened in the present. Everybody does this, and it is certainly not specific to science. In one study, several women were asked to choose their favorite pair of nylon stockings from a group of twelve. Then, after they had made their selections, researchers asked them to explain their choices. Among the explanations, texture, feel, and color were the most popular. However, all of the stockings were in fact identical. The women were being sincere – they truly believed that what they were saying made sense – but like the scientific method, they simply made up reasons in the present for something that happened in the past – and that’s the problem.
The more plausible explanation is that we don’t know how or why a scientific discovery is made; we know that it was made, but that does not mean we know how it happened. Our stories are all after-the-fact constructions that make it seem as if something like the scientific method was followed. But the reality of science is that there is an extreme amount of chance and randomness that is involved, and because we have such a propensity to decorate the past with phony cause and effect stories, it never gets reported.
So when we consider that many scientific discoveries are accidental, and that we have a strong tendency to make up reasons to explain how and why they happened, it seems that we are justified to call B.S. on the scientific method. But the purpose here is not to put down your 8th grade science teacher; it is to convince you that the most important part of science is to not necessarily focus on an end goal and all the steps that precede it, but rather, as Max Planck would have suggested, to ignore the end goal and just start experimenting, testing, and most importantly, doing.