From: Ben Goertzel (ben@goertzel.org)
Date: Fri Feb 20 2004 - 08:08:03 MST
Rafal,
If I understand you correctly, you now seem to be saying something like
"Create goals, and rules that, if followed, will lead to the achievement of
these goals"
or
"Create goals, and rules that, if followed, will lead to the achievement of
these goals, with as few side-effects as possible."
I understand that the goals and rules are typically presented in a combined
format rather than separately, but I'm not sure it does harm to separate
them, and it provides some conceptual simplification.
I agree that these statements are more meta-ethical than your previous
proposal, which in this language was basically like
"Create goal-rule systems that will be accepted"
Your new statement is indeed more general, and is inclusive of the previous
one. "Will be accepted" seems to add a very substantial additional
condition to your new, very abstract meta-ethic.
Because, consider the "ethical" system:
DESTROY ALL LIVING BEINGS BY
1) FIRST STUDYING ALL OTHER LIVING BEINGS SCIENTIFICALLY TO DETERMINE HOW TO
KILL THEM MOST EFFECTIVELY AND AT LOWEST RISK, AND THEN KILLING THEM
2) FINALLY, KILLING ONESELF
This posits a goal and also some rules for how to achieve the goal. It is
rational and consistent. It obeys your new, more abstract meta-ethic.
However, in practice, it fails your former, more concrete almost-meta-ethic,
because at least among MOST OF the sentient beings I know, it is unlikely to
be accepted. (Now and then various psychopaths have of course accepted this
"ethic").
So, to me, by further abstracting your meta-ethic, you have moved from
-- a very abstract formulation of "the good" ["Create goal-rule systems that
will be accepted"]
to
-- a very abstract formulation of the general process of goal-seeking
Seeking the good is a special case of goal-seeking in general....
I think your previous formulation succeeded in getting about as abstract as
"ethics" can get. Your new formulation is *so* abstract it seems more
"meta" than "ethical" ;-)
-- Ben G
> -----Original Message-----
> From: owner-sl4@sl4.org [mailto:owner-sl4@sl4.org]On Behalf Of Rafal
> Smigrodzki
> Sent: Thursday, February 19, 2004 5:17 PM
> To: sl4@sl4.org
> Subject: RE: Ethical theories
>
>
> Ben wrote:
>
> >Rafal wrote
> >>
> >> I think we could begin by making the metaethical statement
> >> "Formulate rules which will be accepted" (although this statement is
> >> actually a high-level link in a very long-term recursive mental
> >> process, rather than a starting logical premise).
> >
> > That's interesting. It's a little deeper than it seems at first, and
> > I need to think about it more.
> >
> > At first it seems a pure triviality, but then you realize what the
> > preconditions are, in order for the statement to be meaningful. For
> > "be accepted" to be meaningful, one needs to assume there is some
> > mind or community of minds that has the intelligence and the freedom
> > to accept or to not accept. So one is implicitly assuming the
> > existence of mind and freedom. So your rule is really equivalent to
> >
> > "Ensure that one or more minds with some form of volition exist, and
> > then formulate rules that these minds will 'freely' choose to accept"
>
> ### You are close to getting to the bottom of the issue here, but
> let me try
> to reformulate the initial meta-ethical statement. As you point out, this
> statement is actually applicable only to ethical systems professed by
> creatures interested in survival - but creatures which don't care about
> their own lives can have ethical systems, too. Let me then make a
> hopefully
> more general meta-ethical statement - "Formulate rules that make
> themselves
> into accepted rules, or, make themselves come true" (cause the
> existence of
> states of the universe, including conscious states, in agreement
> with goals
> stated in the rules). Or "Formulate rules which, if applied, will as their
> outcomes have the goals explicitly understood to be inherent in these
> rules". Or "Do not formulate rules which have outcomes *opposite* to
> intended". If the goal of the rule is to have a "good" outcome, where
> goodness is defined within the rule itself, then only rules which
> have good
> outcomes are good rules, and ethical systems which have good outcomes are
> worth considering. Ethical systems which by their very structure have
> results opposite to or uncorrelated with goals of these systems
> would appear
> to be inferior to those which produce intended outcomes, because the very
> essence of ethics is to define desired outcomes, no matter what they
> actually are. I think that this is the basic meta-ethical statement we can
> make, essentially demanding rationality in ethics.
>
> >From the demand for rationality in ethics one can derive further
> meta-ethical statements. Need for computability - a system which does not
> provide rules sufficient to compute desirability of concrete
> actions open to
> decision-makers (such as the system consisting of the sole statement "Be
> good"), is useless, uncorrelated with outcomes. Internal consistency - the
> system should not make contradictory recommendations for a single
> situation.
> Wide applicability - a system that guides only in a few situations is less
> useful (less correlated with outcomes) than a system applying everywhere.
> Stability under changes of input - systems which totally change
> recommendations after minor changes in inputs are likely to be affected by
> random misinformation and therefore uncorrelated with outcomes. I think
> similar points were made in this thread, sorry for repetition.
>
> All these considerations seem at first approximation to be independent of
> the content of ethical systems, but depend on epistemological features of
> existing minds - which in turn *are* linked to ethics via the shared
> physical environment which caused both our desires and our truth-finding
> faculties to develop. This represents a bit of circularity between ethics,
> and epistemology, but I don't think that such circularity would invalidate
> the meta-ethical statements - it merely makes them contingent on
> the current
> state of our (physical) truth-finding capabilities - but,
> everything we say
> shares this feature.
>
> ---------------------------------
> >
> > If we define happiness_* (one variant of the vague notion of
> > "happiness") as "the state of mind a volitional agent assumes when
> > it's obtained what it wants", then your rule is really equivalent to
> >
> > "Ensure that one or more minds with some form of volition exist, and
> > then formulate rules that these minds will 'freely' choose to accept,
> > because they assess that accepting these rules will bring them an
> > acceptable level of happiness_*"
> >
> > My point in tautologously unfolding your rule in this way, is to show
> > that (as you obviously realize) it contains more than it might at
> > first appear to...
> >
> > However, the shortcoming it has, is that it doesn't protect against
> > minds being stupid and self-delusional. Volitional agents may accept
> > something even if it's bad for them in many senses. (This is because
> > happiness_* is not the only meaningful sense of happiness).
>
> ### Well, as I mentioned above I wanted to say something even
> less dependent
> on our current structure of volition, which for most humans contains a
> desire to exist. I hope that the reworked statement is more general, and
> then it wouldn't entail the need for continued existence of minds
> espousing
> a given ethics, much less the specific content of joyousness or growth. I
> understand that this makes it even less intuitively compelling that my
> initial statement, but it is more meta-ethical.
>
> Rafal
>
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:45 MDT