Re: Ethics as Science

From: Technotranscendence (
Date: Sat Mar 11 2000 - 01:26:17 MST

On Friday, March 10, 2000 5:53 AM Greg Burch wrote:
> I'm very far behind in my correspondence and I know that Robin has said he
> has to drop out of this discussion for work reasons. I've had time to
> read this interesting thread quickly and so make only this one comment
> the peanut gallery: I don't see Robin's points addressing the fundamental
> "is-ought" gap that I think Dan is talking about.

We could discuss Rand's way of bridging the gap, i.e., by looking for what
facts gives rise to values. After all, it would appear only certain kinds
of beings have or pursue values and only under certain circumstances. That
might be better than wondering what someone could possibly believe, etc. (I
haven't followed this discussion too closely either. Hard to do with so
many competing things on and off the list.)

> To me, it's one thing (and
> a very important and good thing) to come up with increasingly accurate
> descriptive and predictive models of behavior,

It would also be nice to see actual models of human behavior that are highly
predictive, especially ones that can not only predict, say, habits or
genetic stuff, but what the effects of a given change will be. Granted, to
some extent we all have this. We interact with other people and expect
certain ranges of reactions from them. We even expect certain norms in film
and fiction.

> and quite another to develop
> PRESCRIPTIVE systems of analysis. As I've written here before, I do not
> any fundamental inconsistency between scientific materialism and
> "determinism" on the one hand and the notion that sentient beings are
> actors capable of making (and required to make) moral choices. In brutal
> summary, my reconciliation of these two notions derives from complexity
> theory: above a certain level of complexity, mental systems cannot be
> accurately enough modeled in practical terms to completely predict their
> behavior, requiring such systems to employ generalizing heuristics that
> satisfactorily onto a traditional notion of the realm of "ethics".

There are a few problems here. One, I question whether materialism is or
can be scientific. Science is typically taken to be a method of gaining
knowledge or knowledge gained via that method. Materialism is an
ontological position. I'm not claiming the twain never doth meet, just that
one does not imply the other.

Two, the problem for morality of determinism is not about prediction per se,
but about responsibility and choice. To wit, if an alleged actor cannot
choose any of her actions, then the typical -- and, to me, correct -- view
is that she's not responsible for her actions. In other words, we can't
praise or blame her for doing right or wrong, good or ill, this or that.
But that does not exhaust morality either. After all, intentions as well as
actions and results have to come into play.

Three, the complexity argument doesn't resolve the issue the way Greg wants
it to be resolved -- with all sides being right and everything taken up into
to some greater whole. Instead, if a system is deterministic, it is
deterministic, whether to itself it cannot be modeled or not. (I would also
submit that when we model our and other people's actions, we need not be
complete or total -- any more than in trying to predict where to point the
cameras on the Galileo probe presumes a complete knowledge of celestial
happenings. But that's what Greg meant by "generalizing heuristics,"

That a system is complex does not clear up everything either. After all,
paradigmatic complex systems tend to magnify small pertubations. That being
so, imagine a determined determinist who despite his scrutiny overlooks one
small disturbance that happens to be completely undetermined. Soon, that
disturbance might affect the whole system. What I'm saying is that
complexity should make us a little humble -- though we should not retreat
before it like some savage before a graven idol.:)

Too tired to write more!

Daniel Ust
    Go Euchner! at:

This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:04:52 MDT