Like I mentioned elsewhere in this thread, the “No Universally Compelling Argument” post you site applies equally well to physical and even mathematical facts (in fact that was what Eliezer was mainly referring to in that post).
In fact, the main point of that sequence is that just because there are no universally compelling arguments doesn’t mean truth doesn’t exist. As Eliezer mentions in where recursive justification hits bottom:
Now, one lesson you might derive from this, is “Don’t be born with a stupid prior.” This is an amazingly helpful principle on many real-world problems, but I doubt it will satisfy philosophers.
A formal proof is still a proof though, although nothing mandates that a listener must accept it. A mind can very well contain an absolute dismissal mechanism or optimize for something other than correctness.
We can understand what sort of assumptions we’re making when we derive information from mathematical axioms, or the axioms of induction, and how further information follows from that. But what assumptions are we making that would allow us to extrapolate absolute moral facts? Does our process give us any way to distinguish them from preferences?
And if he gave a true moral argument you would have to accept it?
How would you distinguish a true argument from a merely persuasive one?
Like I mentioned elsewhere in this thread, the “No Universally Compelling Argument” post you site applies equally well to physical and even mathematical facts (in fact that was what Eliezer was mainly referring to in that post).
In fact, the main point of that sequence is that just because there are no universally compelling arguments doesn’t mean truth doesn’t exist. As Eliezer mentions in where recursive justification hits bottom:
A formal proof is still a proof though, although nothing mandates that a listener must accept it. A mind can very well contain an absolute dismissal mechanism or optimize for something other than correctness.
We can understand what sort of assumptions we’re making when we derive information from mathematical axioms, or the axioms of induction, and how further information follows from that. But what assumptions are we making that would allow us to extrapolate absolute moral facts? Does our process give us any way to distinguish them from preferences?