Models and Truth

Models and Truth

I learned an importan lesson long ago in physics class.

On the blackboard, the teacher drew a picture of an atom similar to the one below:

He then asked if that was the correct model of the atom. We, being know-it-all college students, said no.

So he drew up an orbital model we'd seen in discussing chemical reactions. Again- correct? Again - no.

And so on.

After a few different versions, he stopped, and said we were all wrong. He then explained why.

None of the models on the board were truly atoms. They were maps, symbols, models. They were all, in one way or another, useful ways to think about atoms.

The map was not the territory, and the usefulness of the map depended on how easily it got you from point A to point B.

Much like subway maps are not scaled accurately, because the exact relative positioning of the lines is nowhere near as important for most people than how to navigate the connections.

These models of the atom may not have been completely true, but they reflected an aspect of the truth that let you work with them reliably for certain tasks the model is suited for.

And in a way, that makes them far more true than many scientific theories.

But, science(!!!)?

Let me introduce you to a tweet thread:

This is such a powerful meme but it's completely wrong. The Wright brothers & a century of airplane builders were engineers. Scientists first dismissed flight as impossible even after it happened, then made up a bunch of irrelevant equations to pretend to explain how it happened.

Everything that matters to our modern life was built by engineers and workers who got their hands dirty. Scientists sat in cushy universities writing textbooks after the fact indoctrinating generations to think it was their post-hoc explanations that built things.

In short, anytime you hear scientists explain something, prepend it with "behaves as if".

When you talk about science outside of "things scientists do" or the nigh-religion of teh "I fucking love science" crowd, you get the scientific method, and the accumulated body of knowledge where we know "when you do this, that is the result." The theories we devlop are models that we piece together that says "it behaves as if". In short, "when I do this to the thing, it behaves as if this process happens, resulting in that."

It's a dressed-up just-so-story that, if it to have any value, can be used to correctly predict other results if the starting conditions are changed.

In short, a model of how things work. A useful tool in understanding a system and predicting results.

The thing is, we've been doing things we know work, for millennia, without understanding at all why they work. At a very fundamental level, given the ongoing search for dark matter, strings, and other answers to the nature of gravity, we still do not.

And it doesn't change the fact that engineers and craftsmen have, by trial and error, come up with ways to reliably and repeatably build, construct, and fashion, with only the barest hints of why things work the way they do, or the internal processes involved.