From: Perry E. Metzger (firstname.lastname@example.org)
Date: Sun Jan 04 2004 - 18:30:03 MST
"Ben Goertzel" <email@example.com> writes:
> Perry Metzger wrote:
>> Why do we torture animals to produce tastier food? Because they can't
>> fight back and we like tasty food. Is this moral or immoral? I unask
>> the question -- it is meaningless.
> Perry, it's easy to argue that there is no objective morality, but so what?
> It's easy to argue that there's no objective *anything*.
Indeed, that's what pan-critical rationalism is all about. You test
everything, and keep testing.
Now, I have a hypothesis that this hammer in my hand is an
illusion. An easy way to try to falsify that will be to swing it down
on my hand and see if it hurts or not. ("Ow!")
It is of course possible -- even perhaps likely -- that we're all in a
simulation of some sort. However, absent a way to test that hypothesis
I leave it as an interesting idea that I can ignore from the point of
view of daily life.
Now, we are presented with this idea of an Absolute Morality. A
fascinating concept -- but sadly, non-falsifiable. Indeed, I'm told
about dozens of Absolute Moralities, all of which claim to be *the*
Absolute Morality and all of which seem to have this same property of
non-falsifiability. I therefore see it as a fine class of ideas to
ignore for the present, no matter how interesting they may be.
> Sure, to argue that killing cows is worse than killing bugs but better than
> killing humans, you need to begin from some assumption, such as the
> assumption that causing pain to sentient beings is bad.
But why should I make such an assumption? A Jain would not make it --
he would argue killing the cows or bugs is something to be avoided,
even at the cost of your own life.
> Without making SOME assumption -- however abstract -- as to "basic
> morality" one can't have a meaningful moral discussion.
Correct! And indeed, I would fully argue that one can't have a
meaningful moral discussion and that we should abandon the attempt,
and with it the attempt to define a "Friendly" AI.
> All discussions of morality must proceed within some common
> framework of moral understanding, and hence, as you say, morality is
> not objective.
> However, to argue that the helium atom exists you also need to begin from
> some assumptions, such as assumptions about the reliability of laboratory
And I can test such things repeatedly. "Pan-critical rationalism." I
can't test for the "Moral Fluid" even once -- it seems as ephemeral as
> It's not enough to argue that morality is subjective -- when you poke
> anything hard enough, even a big hard granite rock, it begins to look
> subjective (and I'm not even gonna get started on quantum philosophy right
Kick it and see whether it is fully subjective. :)
> And, I'm not so sure that it is. Much of morality is very subjective and
All of it, actually.
> But, at the core of morality lies the core value of
> compassion, which may perhaps NOT be more subjective than the big hard
> granite rock outside my window....
Why would you argue that?
> Just as a rock has a kind of "partial objectivity" that distinguishes it
> from more thoroughly subjective things like hallucinations and delusions;
> so, perhaps, does compassion?
Can you give evidence for this?
> The question is never whether X has objective, absolute reality or not.
> Nothing does.
Again, most hypotheses are falsifiable. "There is a gorilla in the
room" can be tested by searching for the gorilla. "Compassion is
universal" could be checked by noting the quaint practices of the
Innuit involving placing grandma on the ice floe when she couldn't
care for herself any more. "There Is An Absolute Morality" can't be
> And the point is not to argue that X is subjective, therefore
> meaningless -- that kind of nihilism was already found wanting by
> Dostoevsky, Nietzsche and the whole proto-existentialist crew.
Nietzsche is just pietzsche by me -- why do you find him wanting? :)
Anyway, have a quick read of Karl Popper's "Objective Knowledge" and
then tell me how morals fit into that framework.
-- Perry E. Metzger firstname.lastname@example.org
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:43 MDT