Re: Philosophy, Logic, Formal Systems
Reply #178 –
*What* occurs is accounted for by means of logical analysis of the apparent elements in the system or in the model that is being studied. The "probability" in the process of this analysis is to get the definitions and labels straight and relevant. To get the delimiting definitions wrong is to make the elements in the system inconsistent, "improbable", "unlikely". This is the domain of deduction.
I don't have any serious qualms about this way of putting it… But there really are some who deny the necessity of the model (premises) as a vehicle for deduction! They're hung up on the phrase "knowledge is warranted belief" in conjunction with "science is provisional"… The whole being right, for the wrong reason trope.
It's hours since I should have been a-bed. But, tomorrow (actually, later today!), I'll try to find the paper I most recently read — as an example of what I mean; since I don't think I've made the position clear… Also, because I don't quite understand it myself!
For me, probability is conditional on premises. Always. (And statistics -statistical reasoning, if you will- is a deductive discipline… A form of mathematics.)