When to Trust the Science

P aul Graham is a brilliant investor in new technology companies — or as I think they should be called, “JavaScript businesses.” But he said something the other day that betrays a fundamental misunderstanding about how our world works, one that is surprisingly common among educated people:

This is almost like saying that every time you order McDonald’s french fries expecting that usual, uniform, unctuous salty flavor, you are putting your trust in the McDonald’s corporate shareholders. After all, their investment is, in some way, premised on uniformity as a business model.

It’s become common for a certain type of person to try to enhance the reputation of science and scientists by calling things “science” that are just nature. There are lots of boards on Reddit where you can find images of astonishing natural phenomena — such as dirt devils spinning on a dusty plateau, or majestic storm clouds, or the aurora borealis with captions like, “I love science!”

No, friendo, you are awed by Nature.

Paul Graham is just wrong but wrong in a constructive way. You are not putting your faith in science or scientists when you step on a plane. Scientists were famously divided about how humans could design a flying machine capable of carrying them. Scientists may still be divided about how flight works, if this intriguing article is correct.

When you step on a plane, you are putting faith in human experience, in the laws of nature, in the institution that runs the flights, the governments that regulate them, the engineers who maintain the planes, and the pilots who fly the planes. The involvement of science as a field, or scientists, was discreet. Scientists come up with theoretically more efficient designs of planes and engines, and those are tested in the real world. Once their theoretical designs prove themselves in nature, the trust in the science stops. What then begins is the trust in the constancy of nature’s physical laws and in the incentives that induce airlines to keep their planes in good shape.

The same is true with the can of beans. I once spent some time working in a chemical factory that produced materials put in everything from chewing gum to microchips. Our chemical plant had a few chemists in a separate lab who would verify various reactions and even come up with suggestions for increasing their yield and efficiency in our particular factory.

But their involvement ceased once their work was reduced to a formula and a work process. At that point it was handed over to engineers, operators, and lowly technicians like myself — though even that is dignifying what I did. But what was really amazing is that long-term chemical factory workers, with little theoretical scientific training and only need-to-know operational training in chemistry, would, over decades, begin to develop highly evolved instincts when it came to running large and complex reactions. It was often these workers who would adapt what they learned in other reactions to produce an innovative reaction process that increased our yield and efficiency, often making certain reactions temporarily much more profitable, until the cost savings were eventually passed through the supply chain down to the customer. The vast majority of chemical products we produced were not seen or checked over by “scientists” — but only by experienced workers.

And it’s worth pointing out that Orville and Wilbur Wright were not “scientists” in any way that would be recognized today. They were high-school dropouts — literate, adventuresome, and inventive, but not scientists. The engineering of gliders was something they mastered by trying over and over again. In fact, it took months and, in many places, years for the public to accept the Wright brothers’ claim to have invented and piloted a machine that flies humans. Why? The scientific consensus said it couldn’t be done, especially by bicycle-makers like those guys.

We have to “trust the science” on more theoretical questions where we don’t have sufficient experiences — like on the safety of mRNA vaccines. But science is, in fact, a humble profession that inches along for decades at a time, slumming for good funding. And often, like every other guild, it becomes a prisoner of its own orthodoxies, or institutional model. You’re not putting your trust in science as your plane lifts off — you’re putting it in human experience and institutions, and maybe even taking that leap of faith with a couple of bicycle sellers from the sticks.