= 27d567bc2d48d9f40e9ccc20ea370b15)
Kaj_Sotala comments on Diseased thinking: dissolving questions about disease - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (343)
The law must keep its promises? That doesn't sound particularly Utilitarian, or even particularly consequentialist. Deontologists could endorse the focus on the distinction between behaviors that are responsive to praise/blame and those, like the development of cancer, that are not. Or to put it another way, on the distinction between behaviors that are responsive to talk and those that are not. Here, "talk" includes self-talk, which includes much reasoning.
In this case, the law must "keep its promises" because of what would follow if it turned out that the law didn't actually matter. That's a very consequentialist notion.
The whole thing is a bit of a half-measure. Even if people's actions are predetermined, and they are not morally accountable for them, we need to hang them anyways because without such an incentive, even more people may be billiard-balled into doing the same things.
Of course, this is all quite besides the point. If you swallow billiard-ball determinism hook, line, and sinker, there's really no point in talking about why we do anything, though, as it works out, there's also little point in objecting to people discussing why we do anything, since it's all going to happen just as it happens no matter what.
ETA: I'm referring to the commonly misconceived notion of determinism that thinks free will cannot exist because the universe is "merely" physical. I mean "billiard-ball determinism" as something of a pejorative, not as an accurate model of how the universe really works. I'm not claiming that a deterministic universe is incompatible with free will; indeed, I believe the opposite.
I'm pretty sure this is wrong. A billiard-ball world would still contain reasons and morals.
Imagine a perfectly deterministic AI whose sole purpose in life is to push a button that increments a counter. The AI might reason as you did, notice its own determinism, and conclude that pushing the button is pointless because "it's all going to happen just as it happens no matter what". But this is the wrong conclusion to make. Wrong in a precisely definable sense: if we want that button pushed and are building an AI to do it, we don't want the AI to consider such reasoning correct.
Therefore, if you care about your own utility function (which you presumably do), this sort of reasoning is wrong for you too.
I was evidently unclear. When I say "billiard-ball determinism" I mean the caricature of determinism many people think of, the one in which free will is impossible because everything is "merely" physical. If no decision were "free," any evaluative statement is pointless. It would be like water deciding whether or not it is "right" to flow downhill - it doesn't matter what it thinks, it's going to happen.
I agree that this is not an accurate rendition of reality. I just find it amusing that people who do think it's an accurate rendition of reality still find the free-will debate relevant. If there is no free will in that sense, there is no point whatsoever to debating it, nor to discussing morality, because it's a done deal.
I would also add that this example assumes free will - if the AI can't stop pushing the button, it doesn't really matter what it thinks about its merits. If it can, then free will is not meaningless, because it just used it.
I'm not sure what exactly you mean by "can't". Imagine a program that searches for the maximum element of an array. From our perspective there's only one value the program "can" return. But from the program's perspective, before it's scanned the whole array, it "can" return any value. Purely deterministic worlds can still contain agents that search for the best thing to do by using counterfactuals ("I could", "I should"), if these agents don't have complete knowledge of the world and of themselves. The concept of "free will" was pretty well-covered in the sequences.
You're right, but you're not disagreeing with me. My original statement assumed an incorrect model of free will. You are pointing out that a correct model of free will would yield different results. This is not a disputed point.
Imagine you have an AI that is capable of "thinking," but incapable of actually controlling its actions. Its attitude towards its actions is immaterial, so its beliefs about the nature of morality are immaterial. This is essentially compatible with the common misconception of no-free-will-determinism.
My point was that using an incorrect model that decides "there is no free will" is a practical contradiction. Pointing out that a correct model contains free-will-like elements is not at odds with this claim.
Yes, I misunderstood your original point. It seems to be correct. Sorry.
Psychohistorian disagrees that cousin_it was disagreeing with him.
Very cute ;)
Is billiard-ball determinism a particular variant? If so, what does the billiard-ball part add?
I'm just trying to point out that we can agree with a central point of Yvain's post without endorsing consequentialism. For example, Anthony Ellis <pdf> offers a deontological deterrence-based justification of punishment.
The same goes for Holmes's quip, even if in his case it was motivated by consequentialist reasoning. Especially if we take "your act was inevitable for you" to be an (overblown) restatement of the simple fact of causal determination of action.
Oh, right. Yeah, sure - I agree with that.