From: Ben Goertzel (ben@goertzel.org)
Date: Sat Jan 31 2004 - 12:59:14 MST
Hi,
I've been thinking a lot about ethical theory, and I'm wondering if anyone
knows of an ethical theory that has the same kind of structure as Imre
Lakatos's theory of "research programmes" (in the philosophy of science).
What I'm thinking of would be a theory of "ethical systems" rather than
ethical acts. It would agree that any one act may be ethical or not
depending on the ethical system within which you view the act. But it would
give some high-level criteria for judging ethical systems as wholes. (These
criteria would NOT be of the form "An ethical system is good if it includes
particular rule X, or rules of form Y, etc" -- rather, they would pertain to
the dynamics of growth and change of the ethical system, or the abstract
structure of the ethical system.)
I am working out an ethical theory of this nature myself (slowly, in my
spare time), but I am wondering if a theory of this nature already exists.
Any references anyone has, would be appreciated.
Let me elaborate my line of thinking a little further...
One key point is that it's very hard to evaluate one act as ethical vs.
unethical. The evaluation of an act's ethicalness occurs within some
system, and has to be understood in context. This is similar to how it's
very hard to evaluate one scientific micro-theory, in isolate, as true or
false; the evaluation takes place within some system (some "research
programme") and has to be understood in this context. So the best level to
assess quality, in the philosophy of science, is on the level of research
programmes not small individual theories. Similarly in ethics, the best
level to assess quality is probably the "ethical system" a person or group
follows, not the individual act. In philosophy of science there is some
knowledge about what constitutes a "good" ("progressive" in Lakatos's terms)
research programme, e.g.
-- it makes dramatic, surprising predictions that turn out to be true
-- in most cases, it accounts for new data without requiring highly
complicated workarounds and modifications to its core concepts
-- etc.
These qualities of progressive research programmes can be used to compare
even research programs that are "incommensurable" like classical vs. quantum
physics, or vitalistic vs. molecular biology.
So I was wondering what qualities there are that distinguish a good ethical
system from a bad one. Is there a similar list of properties characterizing
a "progressive" ethical system, which can be used to compare even
"incommensurable" ethical systems?
Clearly, some ethical systems have a badness that's tied in with regressive
(bad) scientific or quasiscientific research programmes [this could be said
about Marxist ethics, National Socialist ethics, etc.].
Habermas's approach suggests that good ethical systems will have the
property that most people, if given the opportunity to freely express their
opinion in an appropriate situation, living in a society that imposes the
ethical system will agree that the ethical system is a good idea. This kind
of "democracy on the level of ethical systems" is more what we have in
modern representative democracy than a democracy on the level of ethical
acts; because with the exception of super-big issues like abortion,
government makes ethical decisions without direct input from the population.
So we elect our rulers partly because we accept their ethical systems. One
problem with Habermas's approach is that he seems to underestimate the
propensity of people to delude themselves and each other (of course, this is
a general weakness of democracy)... but I haven't read him that seriously,
so I'm not sure how he deals with this.
Another way of looking at ethics is in terms of the psychological effect of
an ethical system on the person holding the ethical system... this ties in
with the Buddhist notion that one reason for being compassionate is because
doing so makes you a better person (because it decreases the illusory
boundary between self and other, which in Buddhist thought is one thing
holding us back from enlightenment). Graves' ethical theory is along these
lines; he ties ethical systems in with Maslow's hierarchy of needs and says
that people necessarily have different ethical systems depending on where
they stand in the hierarchy of needs...
So these are three characteristics of good ethical systems, then
1) tied in with progressive rather than regressive research programmes
2) democratically supportable (a la Habermas)
3) tend to lead to positive psychological growth for the person holding the
system
Maybe that's all one can say on the subject, but I was hoping for something
more interesting ;-)
What started me thinking about this was some futuristic speculation about
the ethics of superhuman AI's.... I assume a superhuman AI will develop a
different ethical system than any human system, and it would be interesting
to have some way to think about which superhuman-AI ethical systems are good
versus bad...
Ah well.. back to work..
-- Ben
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:45 MDT