Saturday, October 01, 2022
in a deterministic world, is there probability?
I think i might have heard something like this mentioned on Sean Carrolls podcast. It is something that has been working away at me, somewhere in the background.
I guess that schrodinger's equation is rolling the dice, and its collapse is the measurement. It gets a bit more dodgy in the many worlds scenario.
I imagine that what we are doing is constructing a model of something, and then judging how good the model is. In particular we are constructing a probabalistic model.
It is tempting to think that the 'something' is itself probabalistic, and we are finding it. But this is not something that is being tested. Indeed it could be tricky to show something created by man is probabalistic or not.
So in someway probabalistic models are entirely man made, and just an example of modelling tool kit. For example, prime numbers really are totally deterministic, but probabilty models may help us imagine the basic structures and behaviours, a little bit.
I still need to think about this. It makes me think that the bayesian approach is conceptually less problematic, particularly because it is not constructing anything, but rather refining something give it. You cannot contruct anything from nothing.
Martin
Martin
