Dan74 wrote: ↑Sun Nov 20, 2022 5:15 pm
I think the belief that the sun will rise tomorrow is justified by our understanding of the processes inside the sun and the physical laws that underpin them, rather than simply basing it on past experience. We do, for instance, know that stars can go supernova and why.
This just moves the goalpost IMO. How do we arrive at the theories of what happens in the sun and the physical laws? I assume you will say its to do with induction again?
Basically, there is a lot more going on than inductive reasoning.
Such as? There are assumptions in science too, such as matter and causality. Assumptions though don't give us knowledge of universal laws.
There are, as I've tried to state, underlying axioms about the stable laws underpinning the workings of the universe, and a great deal of deduction.
As I said, yes. There are assumptions in science. Science has to assume certain things, in order to do science.
As to how the laws are arrived at and whether they are deduced, induced or discovered, there is a bit of A and a bit of B (and C), but regardless, countless of pieces of evidence serve to verify their validity. So as Sam pointed out, I think, when they are verified through weighty evidence, it is rational to take them as true.
Verificationism is pretty much dead, sorry to say. Not even those who proposed it argued for it, in the end.
You seem to insist that the only rational belief is one that is shown purely through the use of logic. In an empirical world, where evidence is always about particulars, no general belief would then be "rational" in your view. Even particulars can be doubted, since we only have evidence of our sense and they could be mistaken, tricked, etc. We are left with Descartes dilemma. The only thing we can reliably conclude is that we reason and therefore exist. That's fine if you want to set the bar that high. I am happy to take a pragmatic approach. I happen to believe that it's more rational, since it tallies better with our experience.
My position is that induction, as of yet, can't be justified. Deduction can be justified, based on the language and meaning of words that we use. Deduction however does not give us access to reality. It doesn't establish matter, causality, the strong and weak nuclear forces and so on. It gives us our best theory about how the world is, but this is still limited to concepts. How to bridge the gap between the words we use and their referents is, as far as I can see, an impossible bridge to build.
In terms of statistical evidence, as it is often used in life sciences, this is again solidly based on logic, I'd say. To use a simple example. We know that if we toss a fair coin 1000 times, we are 99.9% sure that we will get between 450-550 "Heads". So if we toss a coin 1000 times and get 600 heads, we can quite reliably conclude that the coin is biased toward "Heads". This is the essence of hypothesis testing, as you know and it is based on probability theory, which is solid logic. Its applicability to a given question is another matter. Sometimes assumptions are made that cannot easily be checked and then the conclusions are in doubt. But other times, we can be quite sure of our conclusions, even if there is a tiny chance that they were just freaky 1000 tosses. It would be rational, in my pragmatic view, to adopt the conclusion that is heavily supported by the evidence than remain agnostic, as if there is no evidence either way.
In order to establish if there is a statistical relationship within data or if the relationship is just random chance, we make use of a P value being set at 0.05. This however is an arbitrary cut off. It's useful though, as a value, in order to argue that there is a significant relationship being observed. It's pragmatic, sure. Induction too is pragmatic. As pattern seeking animals, we instinctively think in terms of induction, every minute of every day. Does that instinct grant knowledge? Does it establish a world of matter, blackholes and universes? Not as far as I can see. Deduction is better in that it can be justified, but as I have said that too doesn't grant us a world of such things. It gives us what is conventionally true, like how the use of "I am" is conventionally true and useful to use.