It’s fairly common, at least in America, for kids to learn that there is a “scientific method” which tends to look something like:

1. Observation
2. Hypothesis
3. Experiment
4. Go back to 1.

It varies; there is often more detail. In general it’s part of the myth that there was a “scientific revolution” in which at some point people began to study the natural world in a radically different way than anyone had before. I believe (though am not certain) that this myth was propaganda during the Enlightenment, which was a philosophical movement primarily characterized by being a propagandistic movement. (Who do you think gave it the name “The Enlightenment”?)

In truth, people have been studying the natural world for thousands of years, and they’ve done it in much the same way all that time. There used to be less money in it, of course, but in broad strokes it hasn’t changed all that much.

So if that’s the case, why did Science suddenly get so much better in the last few hundred years, I hear people ask. Good question. It has a good answer, though.

Accurate measurement.

Suppose you want to measure how fast objects fall. Now suppose that the only time-keeping device you have is the rate at which a volume of sand (or water) falls through a restricted opening. (I.e. your best stopwatch is an hour glass). How accurately do you think that you’ll be able to write the formula for it? How accurately can you test that in experimentation?

To give you an idea, in physics class in high school we did an experiment where we had an electronic device that let long, thin paper go through it and it burned a mark onto the paper exactly ten times per second, with high precision. We then attached a weight to one end of the paper and dropped the weight. It was then very simple to calculate the acceleration due to gravity, since we just had to accurately measure the distance between the burn marks.

The groups in class got values between 2.8m/s and 7.4m/s (it’s been 25 years, so I might be a little off, but those are approximately correct). For reference, the correct answer, albeit in a vacuum while we were in air, is 9.8m/s.

The point being: until the invention of the mechanical watch, the high precision measurement of accurate time was not really possible. It took people a while to think of that.

It was a medieval invention, by the way. Well, not hyper-precise clocks, but the technology needed to do it. Clocks powered by falling weights were common during the high medieval time period, and the earliest existing spring driven clock was given to Phillip the Good, Duke of Burgundy, in 1430.

Another incredibly important invention for accurate measurement was the telescope. These were first invented in 1608, and spread like wildfire because they were basically just variations of eyeglasses (the first inventer, Hans Lippershey, was an eyeglass maker). Eyeglasses were another medieval invention, by the way.

And if you trace the history of science in any detail, you will discover that its advances were mostly due not to the magical properties of a method of investigation, but to increasing precision in the ability to measure things and make observations of things we cannot normally observe (e.g. the microscope).

That’s not to say that literally nothing changed; there have been shifts in emphasis, as well as the creation of an entire type of career which gives an enormous number of people the leisure to make observations and the money with which to pay for the tools to make these observations. But that’s economics, not a method.

One could try to argue that mathematical physics was something of a revolution, but it wasn’t, really. Astronomers had mathematical models of things they didn’t actually know the nature of nor inquire into since the time of Ptolemy. It’s really increasingly accurate measurements which allow the mathematicization of physics.

The other thing to notice is that anywhere that taking accurate measurements of what we actually want to measure is prohibitively difficult or expensive, the science in those fields tends to be garbage. More specifically, it tends to be the sort of garbage science commonly called cargo cult science. People go through the motions of doing science without actually doing science. What that means, specifically, is that people take measurements of something and pretend it’s measurements of the things that they actually want to measure.

We want to know what eating a lot of red meat does to people’s health over the long term. Unfortunately, no one has the budget to put a large group of people into cages for 50 years and feed them controlled diets while keeping out confounding variables like stress, lifestyle, etc.—and you couldn’t get this past an ethics review board even if you had the budget for it. So what do nutrition researchers who want to measure this do? They give people surveys asking them what they ate over the last 20 years.

Hey, it looks like science.

If you don’t look to closely.