= 27d567bc2d48d9f40e9ccc20ea370b15)
Stuart_Armstrong comments on Questions for Moral Realists - 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 (110)
Sorry, I thought you already understood why wanting can be wrong.
Example 1: imagine a person named Eliezer walks to an ice cream stand, and picks a new flavor X. Eliezer wants to try the flavor X of ice cream. Eliezer buys it and eats it. The taste is awful and Eliezer vomits it. Eliezer concludes that wanting can be wrong and that it is different from liking in this sense.
Example 2: imagine Eliezer watched a movie in which some homophobic gangsters go about killing homosexuals. Eliezer gets inspired and wants to kill homosexuals too, so he picks a knife and finds a nice looking young man and prepares to torture and kill him. Eliezer looks at the muscular body of the young man, and starts to feel homosexual urges and desires, and instead he makes love with the homosexual young man. Eliezer concludes that he wanted something wrong and that he had been a bigot and homosexual all along, liking men, but not wanting to kill them.
I understand why those examples are wrong. Because I have certain beliefs (broadly, but not universally, shared). But I don't see how any of those beliefs can be logically deduced.
Quite a lot follows from "positive conscious experiences are intrinsically valuable", but that axiom won't be accepted unless you already partially agree with it anyway.
I don't think that someone can disagree with it (good conscious feelings are intrinsically good; bad conscious feelings are intrinsically bad), because it would be akin to disagreeing that, for instance, the color green feels greenish. Do you disagree with it?
Can you elaborate? I don't understand... Many valid wants or beliefs can be ultimately reduced as to good and bad feelings, in the present or future, for oneself or for others, as instrumental values, such as peace, learning, curiosity, love, security, longevity, health, science...
I do disagree with it! :-) Here is what I agree with:
I'll also agree that preferences often (but not always) track the positive or negative conscious experiences of that human. That human impressions of good and bad sometimes (but not always) track positive or negative conscious experiences of humans in general, at least approximately.
But I don't see any grounds for saying "positive conscious experiences are intrinsically (or logically) good". That seems to be putting in far to many extra connotations, and moving far beyond the facts we know.
I agree with what you agree with.
Did you read my article Arguments against the Orthogonality Thesis?
I think that the argument for the intrinsic value (goodness or badness) of conscious feelings goes like this:
Conscious experiences are real, and are the most certain data about the world, because they are directly accessible, and don't depend on inference, unlike the external world as we perceive it. It would not be possible to dismiss conscious experiences as unreal, inferring that they not be part of the external world, since they are more certain than the external world is. The external world could be an illusion, and we could be living inside a simulated virtual world, in an underlying universe that be alien and with different physical laws.
Even though conscious experiences are representations (sometimes of external physical states, sometimes of abstract internal states), apart from what they represent they do exist in themselves as real phenomena (likely physical).
Conscious experiences can be felt as intrinsically neutral, good, or bad in value, sometimes intensely so. For example, the bad value of having deep surgery without anesthesia is felt as intrinsically and intensely bad, and this badness is a real occurrence in the world. Likewise, an experience of extreme success or pleasure is intrinsically felt as good, and this goodness is a real occurrence in the world.
Ethical value is, by definition, what is good and what is bad. We have directly accessible data of occurrences of intrinsic goodness and badness. They are ethical value.
Of course!
-> Likewise, an experience of extreme success or pleasure is often intrinsically felt as good, and this feeling of goodness is a real occurrence in the world.
And that renders the 4th point moot - your extra axiom (the one that goes from "is" to "ought") is "feelings of goodness are actually goodness". I slightly disagree with that on a personal moral level, and entirely disagree with the assertion that it's a logical transition.
This is a relevant discussion in another thread, by the way:
http://lesswrong.com/lw/gu1/decision_theory_faq/8lt9?context=3
Could you explain more at length for me?
The feeling of badness is something bad (imagine yourself or someone being tortured and tell me it's not bad), and it is a real occurrence, because conscious contents are real occurrences. It is then a bad occurrence. A bad occurrence must be a bad ethical value. All this is data, since conscious perceptions have a directly accessible nature, they are "is", and the "ought" is part of the definition of ethical value, that what is good ought to be promoted, and what is bad ought to be avoided.
This does not mean that we should seek direct good and avoid direct bad on the immediate present, such as making parties to no end, but it means that we should seek it in the present and the future, seeking indirect values such as working, learning, promoting peace and equality, so that the future, even in the longest-term, will have direct value.
(To the anonymous users who down-voted this, do me the favor of posting a comment saying why you disagree, if you are sure that you are right and I am wrong, otherwise it's just rudeness, the down-vote should be used as a censoring mechanism for inappropriate posts rather than to express disagreement with a reasonable point of view. I'm using my time to freely explain this as a favor to whoever is reading, and it's a bit insulting and bad mannered to down-vote it).
Why? That's an assertion - it won't convince anyone who doesn't already agree with you. And you're using two meanings of the word "bad" - an unpleasant subjective experience, and badness according to a moral system. Minds in general need not have moral systems, or conversely may lack hedonistic feelings, making the argument incomprehensible to them.
I have a personal moral system that isn't too far removed from the one you're espousing (a bit more emphasise on preference). However, I do not assume that this moral system can be deduced from universal or logical principles, for the reasons stated above. Most humans will have moral systems not too far removed from ours (in the sense of Kolmogorov complexity - there are many human cultural universals, and our moral instincts are generally similar), but this isn't a logical argument for the correctness of something.
If it is a bad occurrence, then the definition of ethics, at least as I see it (or this dictionary, although meaning is not authoritative), is defining what is good and bad (values), as normative ethics, and bringing about good and avoiding bad, as applied ethics. It seems to be a matter of including something in a verbal definition, so it seems to be correct. Moral realism would follow. It is not undesirable, but helpful, since anti-realism implies that our values are not really valuable, but just fiction.
I agree, this would be a special case, of incomplete knowledge about conscious animals. This would be possible for instance in some artificial intelligences, but they might learn about it indirectly by observing animals, humans, and getting contact with human culture in various forms. Otherwise, they might become morally anti-realist.
Could you explain a bit this emphasis on preference?
Which is exactly why I critiqued using the word "bad" for the conscious experiences, using "negative" or "unpleasant", words which describe the conscious experience in a similar way without sneaking in normative claims.
Er, nothing complex - in my ethics, there are cases where preferences trump feelings (eg experience machines) and cases where feelings trump preferences (eg drug users who are very unhappy). That's all I'm saying.