Stathis Papaioannou wrote:
This sort of argument has been raised many times over the centuries, both by rationalists and by their opponents, but it is based the fundamental error of conflating science with ethics. Science deals with matters of fact; it does not comment on whether these facts are good or bad, beautiful or ugly, desirable or undesirable. These latter qualities - values - are necessarily subjective, and lie in the domain of ethics and aesthetics....
Saying that life is worth living, or that you believe it is bad to kill, are simply statements of your values and feelings, and as such are valid independently of any scientific theory.
It may not be an error to equate science and ethics. Science continually moves into new domains.
I'm of the opinion that there is a valid utilitarian theory of co-operating intelligent agent ethics.
"Utilitarian" because the purpose of the ethical principles can be shown to be "group success"
(i.e. emergent-system survival / success in the competition with other potential
variants of emergent intelligent-agent systems that don't include ethical principles as
behaviour guides for their the agents.)
Note the subtlety that the utility NEED not be to an individual agent directly, but may only
accrue to individuals in the group, ON AVERAGE, due to the ethics and moral rules generally
obeyed by the group members, and the consequent "floating of (almost) all boats".
One of the common debates is between ethical/moral relativism versus absolutism.
I call this a confusion due to oversimplification of the issue, rather than a debate.
In this regard, this debate is as silly as the nature vs nurture debate and its influence on,say,
human behaviour, in which the answer is "of course it's a complex feedback loop involving
the interaction of inherited traits and the accidents of life. Duh!" There is no nature vs nurture.
It's always nature AND nurture. Arguing about which is more fundamental is truly unproductive
hair-splitting. We should be researching exactly how the feedback loops work instead.
So completely analogously, with absolute, and relative morals and ethics.
My position is that there are absolute ethical principles and moral rules, but that those
are all general rules, not instantiated rules. (i.e. absolutes in ethics/morals are
all universally quantified rules that apply to general classes of situations and actions.)
Relativism is justified in as far as it is simply debate about how the absolute general
ethical and moral principles should map (do map) onto the current particular situation
at hand. This mapping may not be simple. A single situation can be boundary-scoped
differently, for example, or its agents can be seen as engaging in several different kinds
of acts, with many effects for each act, and the importance to the essence of the situation
of each act and effect can be debated from different perspectives that involve the interests
and knowledge of different agents. So the "single" situation may map validly to several
different instantiations of several ethical principles. And the moral rules applicable to
the situation may be subject then to legitimate debate.
Relativism may also question whether some "moralist" group's absolute moral principles
are general enough, and may argue with some validity that they are not general enough
to be applied without frequent error (and tragedies of injustice).
e.g. "Dont Eat Pork" --> Yeah, whatever
however, "Don't eat the kinds of meat that are often rotten and disease-ridden in our climate, like Pork"
may be a valid moral rule at some historical time and place.
e.g. "Thou shalt not kill." --> Well that's an easy to remember simplification, but a little over simplified and too specific.
How about:
"Minimize the amount of quality-life-years lost in this encounter."
So, "women and children first into the lifeboats. You old geezers are shark-bait."
Or.. "Take out the guy wearing the bomb. Now."
And relativism is also justified in as far as it is the correct observation that
many (most) situations of complex interactions beteen multiple intelligent agents can
be described from multiple perspectives (and/or multiple situation-scope inclusions/exclusions).
A specific situation can be (probably validly) described as co-incident incidences of
the several instances of several different general ethical principles.
A to B
"Our people have lived here from time immemorial. And your grandfathers killed my grandmother.
You are pestilent invaders. Get out or we will have a just war against you."
B to A
"Our people have lived here from time immemorial. And your grandfathers killed my grandmother.
You are pestilent invaders. Get out or we will have a just war against you."
Clearly, it is easy to imagine a situation in which both A and B are factually correct, except perhaps in their
use of the word "just".
Most complex interaction situations requiring application of ethics and moral rules are of this kind,
with competing "goods" or "rights and wrongs".
But it is wrong to suppose that this means there can be no absolute definition of right and wrong.
It's only that only a very general (though clear and simple, at its general level) (and ambiguously mappable)
absolute definition can be constructed, due to the complexities and variations and multiple valid
perspectives and situation-scope-choices in the situations to which we want the concepts to apply.
Examples of general ethical principles: (and their mapping problems)
It is wrong to act against the interests of the group. (But which group, if groups are in conflict?)
It is wrong to reduce the emergent complex order in your neighbourhood
(but if different complex orders compete, how do we choose which to promote?)
While it is alright and perhaps even "good" to compete with other humans, allow the competition to be
fair and limited and governed, in order that "good games" which generate "boat-floating" value for all
or many, can be conducted. (But how much competition, versus how much co-operation, is the
right balance?)
Eric
