From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sdd.hp.com!think.com!Think.COM!moravec Mon May 25 14:04:59 EDT 1992
Article 5607 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!sdd.hp.com!think.com!Think.COM!moravec
>From: moravec@Think.COM (Hans Moravec)
Newsgroups: comp.ai.philosophy
Subject: Re: AI failures
Date: 13 May 1992 01:51:39 GMT
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 134
Distribution: world
Message-ID: <upsnbINNk2c@early-bird.think.com>
References: <MORAVEC.92May10004528@turing.think.com> <1992May12.081742.22060@norton.com>
NNTP-Posting-Host: turing.think.com

In article <1992May12.081742.22060@norton.com>, brian@norton.com (Brian Yoder) writes:
|>  
|> On what basis do you claim that morality is an issue of *social* rules?  
|> Doesn't morality have anything to say to people living on desert islands 
|> all alone?  Things like "Don't be lazy!" and "Don't take foolish risks!".

Society conditions us (especially as children) to act in ways we
probably wouldn't if we grew up in a state of nature.  That conditioning
carries over in later life, even to desert islands.

|> If anything, the fact that morality is a matter of life or 
|> death means that we ought to use our best tools to address the questions at 
|> hand (ie. reason rather than unidentified emotions or traditions).

Sure.  But reasonable people make mistakes, and the bottom line is still
trying things out and seeing how they work.  Marx's reasoning convinced
a lot of people on how to structure a just society.  His conclusions
were tried out.  They didn't work.   (Social) Evolution in action.

|> slavery kept enormous numbers of people from developing their powers
|> fully and resulted 
|> in arrested development of civilization.  Are you saying that because it 
|> existed it must have been good or must have served a purpose (and a 'good' 
|> one at that!)?   Incredible!

I think a strong argument could be made that the early civilizations
depended on slavery the way ours depends on mechanization, and wouldn't
have worked otherwise.  The citizens of Egypt, Greece and Rome had
the leisure time to develop many, many things that remain the basis of our
world because slaves were doing most of the necessary the hard work
to support basic survival.

   > Of course, name calling is an effective tool in political debate,
   > and sways opinions.

|> "Useful" for what?  Deceit?  Do you think that morality is just a con game 
|> for politicians who want to gain power?  You certainly seem to.

	I think moral claims are often used as a political ploy to sway
opinions for or against a government, a cause, a group.  Look how much
mileage generations of US politicians gained by crusading against godless,
immoral, communism.  And how much mileage communist politicians gained by
crusading against evil, exploitative capitalism.

|> Aside from a few primitive reflexes (like the suckling reflex in infants) 
|> human beings HAVE NO INSTINCTS.  Unlike animals, we have conscious control 
|> of what we do.  How do you propose to demonstrate that these "instincts" 
|> exist?  If you could somehow show that instincts (rather than choice) 
|> determined our actions in some context, that would remove that area from 
|> the field of morality, since morality deals specifically with matters of 
|> choice.  If you have no choice about your actions, it makes no sense to 
|> evaluate the morality of that situation.

When we haven't eaten for a while we become hungry because of a
conscious decision to be hungry, to salivate when food appears?
When we reach puberty we suddenly become attracted to the other sex
because we decided to do so the year before?  When someone punches
us in the nose, we become instantly angry because of what we've been
taught?   Mothers hold and protect their babies because of what
they've seen on TV?   Children and adults form peer friendships
because they learned how to in a book?  We enjoy looking at wide open
spaces because we were taught to in school?

Our social structures are built around many more instincts like these,
in a sense perverting them to build something quite unlike the tribal
context where they were initially developed and  tuned.  But like
the biological genes that build the instincts, the
cultural perversions are also inherited, via punishment/reward
conditioning, by example and teaching.  The cultural part evolves
a lot faster than the biological part, though.  For one thing
the contents of a mind can change several times in a lifetime,
but the DNA based biological heritage is frozen in an individual.

|> Why do you think that such behaviors must be instinctual?  Couldn't they be 
|> matters of choice?  Remember though, if there's no choice then there's no 
|> morality involved.

The choice/instinct distinction is very artificial.  As a mechanist
AI type, I think what we at one level perceive as choice is the
interaction of a lot of mechanisms that can be called instinct,
modulated (programmed) to some extent by what we learn.  A given
morality is a set of learned modulations.  Things like, don't
kill the cretin who made a pass at your spouse, even though you feel
an instinctive murderous urge.  Or don't court your neighbor's spouse,
even when instinctive urge suggest otherwise.  A good morality,
like a good set of genes, is a collection of modulations that results
in effective survival (of themselves, and incidentally their hosts).
Since the DNA genes and the cultural heritage reproduce by diferent
means, they don's have exactly the same survival criteria, so there
can be a persistent natural tension between the two.

   > become fine upstanding MORAL citizens.  Exactly what's right and what's
   > wrong keeps changing, 
|> Do you mean that they keep changing because the specific applications keep 
|> changing (eg. there was no need for moral positions on issues like drug 
|> abuse before the drugs were invented), or because society can just choose 
|> to make this or that behavior "moral"?

Some components are more persistent than others, presumably
because they're more broadly useful than others, but morals,
even of most basic things, DO change:
For instance, in the communist world, private ownership of property
was one of the most immoral things there was, the most heinous
of crimes.  It's still illegal by laws on the books.  But, with
the failure of the system built around its rejection, private
ownership has became a good thing, supported by most.   At one time
burning heretics and witches was a good moral thing to do, because it
destroyed only their unimportant mortal bodies, and had a high probability
of causing them to repent, so saving their infinitely more valuable
immortal souls.   Now societies generally see it as immoral.

|> > I'll take E.O. Wilson over your favorite dead humanist windbag any day.
|> 
|> I'm not familiar with Wilson.  What's his story?

Harvard biologist known as the father of sociobiology.  Much
of his life's work is a definitive study of insect societies.
Really outraged the left in the 1970s by applying evolutionary
reasoning to social development in humans.  Was doused with
water by yippies while giving an invited lecture at a AAAS
meeting around 1975.
There was a lot of controversy about his positions (most of
it by some of his left leaning former students like Roger Lewontin)
in the 1970s, but that has faded away and his views are now
part of the biological mainstream.
The message hasn't gotten through to the softer heads
in the humanities, though.

You'd love to hate him.   I'd love to see you hate him.
Like watching Elmer Fudd driven to apoplexy by that cwazy wabbit.

|> -- Brian K. Yoder (brian@norton.com) - Maier's Law:                         

			-- Hans


