Re: The Religion of Arbitrary Value

The Low Golden Willow (
Mon, 10 Feb 1997 23:08:55 -0800 (PST)

On Feb 8, 9:08pm, Eliezer Yudkowsky wrote:
} Subject: The Religion of Arbitrary Value

} Certain folk believe that they can assign to Life whatever meaning they
} choose. That, in other words, value is arbitrary. I contend that this
} is a circular belief system, and one imposed by evolution.

As the entire universe so far seems to be arbitrary, I don't have much
problem with this. Sure, you may be able to argue that we're very
complex thermostats. So what?

} Let me define "circular belief". A circular belief system is one that
} you hold because you hold it, and, if you *stopped* holding it, would
} not pop up again. What atheists hate about religious beliefs is their

And this applies to survival. If I stop valuing my survival, that goal
may not come up again. In which case, I won't be around much longer
either. Of course, one could have other goals, which justify valuing
survival because otherwise those won't get accomplished.

But yes, I can easily see us as machines set to fulfill some purpose.
We hold those values because that is what we are; our values are part of
our definition. We're evolutionarily supposed to value our
reproduction; in fact, we're so complex that a fraction of us have
wandered off and developed new goals, like suriving no matter what, or
having fun, or exploring lots of place, or trying to learn all
knowledge, or realize some concept of a Singularity. Take away those
goals, and you have a different machine, or a broken one which falls

I'm not sure what "meaningful" means for this conversation. Is wanting
to be a polymath meaningful? It's what I am. I could try to justify it
practically -- "knowledge is power, this is good for survival" -- but
that begs the question of survival, and isn't really accurate. I like
to learn for its own sake. I think I also value my survival, out of
stubboness. So I have at least two independent goals, which happen to
support each other somewhat. I can't a priori justify either. It's not
that I haven't thought about it; I have, a fair bit. I haven't gotten
anywhere, and neither has anyone else. This doesn't bother me. Those
it does bother can die of apathy, and leave the rest of us here.

[1] Dies.

} Moreover, I'd like to remind everyone that this issue doesn't seem at
} all Euclid-arbitrary to me. It's not a matter of: Some accept this
} postulate, some don't. At worst it's Godel undecidable, which means

Why not? Actually, your definition of circular systems seems off. Circular
systems are like Escher's waterfall and staircase, where one thing leads
to another over and over. I think what we're talking about here _is_
Euclid arbitrariness. Certain values are imposed by evolution and
upbringing. They aren't justified by their conclusions; they're just
there. Not circular.

} think that the basic postulate - "meaning is arbitrary" - is simply
} WRONG, and what I want everyone to accept is that anybody who starts out
} with this postulate is simply WRONG until they justify it.

If that's what you want, you'll have to work for it. I don't think the
postulate is wrong. Actually, I'm not sure it's terribly well defined
to be either right or wrong.

} How do I reason to it without accepting it as given?

I dunno. I take some things as given. Can you give any systems, of any
complexity, that don't either rely on something else or that aren't
'given'? Something which is self-contained and self-justifying? You
believe the entire universe must be such a thing, but so far that's just
been your affirmation.

Merry part,
-xx- Damien R. Sullivan X-) <*>

"How unfortunate that you should have a reasonable answer, and that I
should be so reasonable as to accept it!"
-- Jane Austen, _Pride and Prejudice_, Elizabeth