apeiron
Gold Member
- 2,138
- 2
bohm2 said:Here’s some interesting arguments against the probabilistic argument presented in original post:
Excellent references for the points made earlier in this thread. Norton skewers Inwagen very nicely.
What Norton calls complete neutrality is of course what Peirce means by vagueness. A realm of pure possibility without any constraints. And so a realm to which the principle of contradiction (ie: of crisp disjunctions) fails to apply.
Bayesian reasoning has to guess at the constraints that might apply to shape pure possibility into some more definite distribution. And where such reasoning goes wrong is when it does not realize this is what it is doing.
It is the same mistake as in trying to apply set theory. Set theory has to presume some global state of constraint to apply to a distribution. Yet vagueness, or neutrality, is something different - it is the truly and profoundly unlimited. There are no constraints by definition. So a larger model of logic is needed to deal with the case - one such as Peircean logic that includes abductive leaps to get things started. And then semiotic constraints where - invoking final cause - it is not local change that is the isssue, but the emergence eventually of limits to change, to the expression of raw possibility.
So Norton provides an argument against all attempts on the "why anything?" question based on conventional probablity approaches - ones that have to already presume constraints on raw possibility. A vagueness doesn't have countable states, not even an infinity of them, as this would already give it something definite, something crisply developed. You have to step back further to a higher level of modelling, one like Peircean semiosis which has that "extra hidden dimension" Norton mentions.
But then of course once armed with the Peircean view, you can start to say something positive about the "why anything?" question.
As I stressed in post #139 - https://www.physicsforums.com/showpost.php?p=3665001&postcount=139 - once you take a view of probability spaces as things that develop, rather than simply exist, then the correct foundational dichotomy of vague~crisp comes into sight. Instead of trying to contrast the likelihood of nothing vs something, we are now talking about the likelihood of the vague yielding the definite. And how that then compares with the universe as we observe its developmental history.
Last edited: