Thread by @skdh

This is a very confused article. Ht @WKCosmo for drawing my attention to it.

aeon.co/essays/why-is-…

@WKCosmo The problem with this article is that it confuses two entirely different notions of simplicity. The first is relative simplicity, aka, Occam's razor. It's where all the stuff about Bayesian inference fits in.

Relative simplicity says: From two theories that explain the same observations, take the simpler one. Occam's razor is a *necessary* criterion for science to work. If it wasn't for it, you'd be allowed to add superfluous assumptions like "and god exists" to your hypothesis.

The article conflates Occam's razor with absolute simplicity, which is an entirely different criterion. That's the belief that the laws of nature should be simple, period. That's a belief, a wish, an aesthetic criterion, a metaphysical assumption -- it's unscientific.

That doesn't mean it's wrong. It just means that there's no reason to think it's right. The fundamental laws of nature might well be more *difficult* than the ones we currently have. That requiring absolute simplicity isn't a useful criterion for theory development is easy to see

It's easy to come up with theories that are simpler than our current ones. A universe without matter, eg. But that doesn't describe what we observe. Absolute simplicity thus eventually can only be scientifically justified if it's "as simple as possible" to describe observations.

And that, well, is Occam's razor again. The bottom line is that it's simply (cough) false to assume that just because Occam's razor is a good scientific criterion we should expect theories to become simpler with more evidence. It's a non sequitur.

Heady stuff, I know, and there's more to say about what that means for multiverse theories, but I'll leave it at this. I explained this all in my book Lost in Math:

amazon.com/Lost-Math-Beau… /end