[hinkley]

This is sort of the argument in Science. In the public discourse you have these conversations about Science being Right or Wrong. Within Science, you have more of this notion of All Models Are Wrong, But Some Are Useful.

We don't know if half of the things we know about physics are right or illusions. What we do know is that the models we have are consistent with the observations, and more importantly that they survive interaction. If I poke it, does the 'illusion' continue, or does it pop like a bubble?

Maybe magnetism doesn't work the way we think it does, but the way we think it works lets me cram millions of dancing lights onto a flat surface and organize them into pictures. And we got to millions of dancing lights from tens of thousands because we kept refining that model down and down into scenarios too small to see with the naked eye.

[andreareina]

I've heard physicists talk about what's real as the set of concepts/objects that are required to describe and predict the world. So imaginary numbers are real (if you believe the qm wave function is real), infinities aren't (but are still very useful).

[LeegleechN]

There are always alternative ways to describe the same phenomena. For example QM could be done with only real numbers (you can emulate imaginary numbers with matrix operations). There is no clear cut notion of what is "required".