Re: ethics is knowable

From: Anders Sandberg (asa@nada.kth.se)
Date: Fri Jun 08 2001 - 06:47:06 MDT


fredagen den 8 juni 2001 10:44 Francois-Rene Rideau wrote:
> >: Anders Sandberg
> >
> > As I see it, ethics is really part of your policy function [...]
>
> Exactly. Actually, I'd say it IS our whole policy function,
> not just the backbone, as you say.

Well, that would be an objectivist way of seeing it.

> But only a backbone
> can be shared between people through objective arguments
> (or said more rigorously: the less universal the ethical rule,
> the less shareable it is). In this, ethics doesn't differ
> from most kinds of knowledge.

Perhaps more importantly, the backbone is more easily algorithmically
compressed and transmitted. I can easily explain general ethical principles,
but when we get down to my ethical-aesthetic policies in regards to potted
plants things get very complicated. It is not the lack of objective arguments
which makes it hard to transfer the rules, but their sheer contingency.

> Sure. But that is in no way specific to ethics.
> Just any practical knowledge is subject to the very same limitations.
> That doesn't mean that they are unknowable, as argued Ben Goertzel, in
> "[foo] is a branch of ethics, i.e. there is no given right or wrong
> answer." Ethics is every bit as knowable as any human-related thing.

Well, one must always ask oneself "what are the fundamental values for this
system" when dealing with ethics. The optimality at achiving value of
different ethical systems with the same values can in principle be compared
objectively. But these values are themselves somewhat arbitrary, although the
big fitness function of living in this universe has selected them to a great
extent.



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:08:01 MST