sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!

Server stats:

537
active users

#induction

1 post1 participant0 posts today

My thinking is that uniformity and disuniformity are the only two options and that one or the other has to be caused and can't be causeless. This is based on my conceptualization of causation as necessitation in a deductive sense. Like how when you have a water molecule you must have an oxygen atom or that when you have up you must have down.

A chicken might expect to get fed like all the other times but instead is butchered that day. A human might expect a coin to land on heads again because it's landed on heads every time before. These are false expectations based on repeated past observations. The first is because they were approaching a threshold they were unaware of. At least that's part of it. The second is because the two only possible outcomes are heads or tails making it 1/2 on average even if they got heads 10 times. This might hold the solution because it tells us why they're false expectations and how to know they are.

But yeah, to simplify and clarify what I've said before I think the problem is that observed repeated correlations aren't enough. We obviously don't know the elements in the necessary relationship like the way we know that when there is a water molecule there must be an oxygen atom. We need to come to know those elements and the necessary relationship itself through imagination and different observations. Then it is known through deduction and is a deductive relationship.

Up to now I reject uniformity to a good degree but do still believe certain things are uniform. BUT I want to be conservative on this claim. For example, C = pi * d is uniform BUT that assumes there isn't logic out there that disputes this. And logic is relative. Although we can say that in our location things have been uniform to a degree up to now, assuming our memories are true and not false. BUT I'd largely adopt relativity in contrast to uniformity.

Observing repeated correlations isn't enough to establish causation, even thought it might be pointing us in the right direction. We need to figure out why one is causing the other if it is. We can figure this out through imagination and different observations. To become aware of that necessary relationship which is deductive in nature. To become aware of what is necessitating what. Something along the lines of if there is a water molecule there must be an atom of oxygen but for causation or necessitation. I say this while being aware that logic is relative to a degree or maybe just relative.

I disagree with Hume. I consider causation to be a deductive relationship that can be known through deduction. If A exists B must exist as a deductive relationship. We can imagine an alleged cause A existing without an alleged effect B but that's only because all we have so far are the observed correlations and thoughts of those observed correlations. We don't yet have the thoughts we need for that deductive relationship of necessitation. Through more observations and imagination we can come to new ideas that establish this relationship as more than a mere observed repeated correlation but a necessitation or causation. That's my thinking anyways.