= 27d567bc2d48d9f40e9ccc20ea370b15)
PhilGoetz comments on Human values differ as much as values can differ - 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 (205)
I recently wrote a long post arguing that your actions specify what your terminal values are, in a way that your thoughts can't. And you commented on it, so I won't repeat the relevant parts here.
click Ok, I think I see now. Forgive me for asking instead of trying to figure this out on my own - finals are looming tomorrow. Morning. And the next morning.
One of two things seem to be going on here. I'll quote Eliezer's metaethical position again, for reference (answering what the meaning of "right" is):
The bolded part is most relevant to my question. Are you agreeing with Eleizer's argument, and just arguing that present terminal values can only be inferred from action? Or are you disagreeing and arguing that present terminal values, which only can be inferred from action, are the terminal values, i.e. the meaning of "right"/"should"? Or is it something else and I'm still confused?
So - first, I don't really believe in terminal values. When I use that term, I'm working within a hypothetical frame (if X is true, then Y); or using it as an approximation.
I said that the only real values an organism has are the values it implements. If you want to have a science of values, and for instance be able to predict what organisms will have what values, you want to work with the behaviors produced, which should follow certain rules; whereas the values that an organism believes it has are different from the values it implements due to accidents of evolution, and working with them will make things less predictable and a science of values more difficult.
Eliezer lists only propositional content as factors to consider. So I think he's not talking about the difficulty of dividing an organism into values and value-implementing infrastructure. He seems to be saying that currently-implemented values are a poor copy of a Platonic ideal which we can extrapolate.
I would be less likely than Eliezer to consider my present values very similar to the "right" values. I think he would either say there are no right values, or that the right values are those extrapolated from your current values in a way that fixes accidents of evolution and flawed cognition and ignorance. But he doesn't have an independent set of values to set up in opposition to your current values. I would, by contrast, feel comfortable saying "existence is better than non-existence", "consciousness has value", and "complexity is good"; those statements override whatever terminal values I have.
I don't really think those statements are in the same category as my terminal values. My terminal values largely concern my own well-being, not what the universe should be like. My preference for good things to happen to me can't be true or false.
I'm not comfortable with using "preference" and "value" interchangeably, either. "Preference" connotes likes: chocolate, classical music, fast cars, social status. My preferences are just lists of adjectives. "Value" connotes something with more structure: justice, liberty. Language treats these things as adjectives, but they aren't primitive adjectives. They are predicates of more than one argument; preferences seem to be predicates of just one argument.
Dude, we both need to stop this and go to sleep.