🚀 go-pugleaf

Hal's Extropians Digest      Sun, 14 Jun 1992       Volume 92 : Issue   0005

    Selected messages from the Extropians list.  To join, send to:
    <extropians-request@gnu.ai.mit.edu>.  Messages from the Extropians
    list may not be cross-posted without permission from the poster.
    Such permission has been obtained for all messages in this Digest.

    Hal Finney
    Extropians Digest Moderator

Today's Topics:
	    Extropy Institute/Exponent/Extropy Information
		   Ethics and objectivists (2 msgs)
	       Ethics & the prisoner's dilemma (5 msgs)
----------------------------------------------------------------------

Date: Wed, 10 Jun 92 15:37:04 PDT
From: uunet!usc.edu!more@nunki.usc.edu (Max More)
To: Extropians@gnu.ai.mit.edu
Subject: Extropy Institute/Exponent/Extropy Information

EXTROPY INSTITUTE
.
Join us on the leading edge of the evolutionary wave as we build a better
future!
.
o Do you want to make the future better than the past?
.
o Do you want to live beyond the normal human limit of 110-120 years?
.
o Are you attracted to innovative, market-oriented solutions to social
problems?
.
o Do you want to be able to explore the universe?
.
o Do you want to augment your cognitive abilities?
.
o Do you want to throw off ingrained barriers to effective thinking and action?
.
o Do you want to continuously improve yourself - physically, intellectually,
morally?
.
o Do you think that technology can help us overcome historical limits to our
abilities and goals?
.
Then the Extropy Institute (ExI) is for you!
.
ExI members receive:
.
EXTROPY: The Journal of Transhumanist Thought
EXPONENT (a bi-monthly newsletter)
Announcements of conferences and social events
Discounts on tapes, videos, lectures, and more
.
        By joining the Extropy Institute you can participate in and support
our projects. As science comes to understand and control the aging process,
stretching healthy human lifespan beyond a century, we will have to re-think
our view of life. Indefinitely living persons will need a worldview effectively
suited to a vastly extended life. Religion cannot provide this worldview, and
humanism is too shortsighted. The Extropian philosophy of life is a
transhumanist philosophy. Extropianism affirms:
.
Boundless Expansion: Seeking to overcome natural limits on our lifespan and
individual and social potential, expansion into space, and acceleration of our
evolutionary development into the transhuman phase -- where human biological
limits will have been overcome.
.
Self-Transformation: A commitment to continual moral, intellectual, and
physical self-improvement, using reason and critical thinking, personal
responsibility, and experimentation.
.
Dynamic Optimism: Positive expectations to fuel dynamic action. Adoption of a
rational, action-based optimism, and a rejection of blind faith and gloomy
pessimism.
.
Intelligent Technology: Applying science and technology to transcend limits
imposed by our biological heritage and environmental  conditions.

Spontaneous Order: Promotion of decentralized,  non-monopolistic social
coordination mechanisms. Fostering of tolerance, long-term planning, and
individual incentives and personal liberty.
.
We aim to educate the public in the uses of computer technologies,
biotechnology, nanotechnology, and psychological means of enhancing human
intellectual, physical, and psychological capacities, and the lengthening of
the human lifespan.
.
To further these values, the Extropy Institute will organize conferences, hold
seminars (also to be available on tape and diskette), and other meetings,
publish both a journal providing in depth analysis, and a newsletter --
Exponent -- providing news and shorter writings, as well as developing
materials for the public and conducting research.
.
Extropy: The Journal of Transhumanist Thought has provided detailed analyses
and discussions since 1988 on topics like Transhumanism and Futurist
Philosophy, Life Extension, Cryonics and Immortalism, Artificial Intelligence,
Supercomputing, Neural-Computer Integration, Personality Uploading, Smart
Drugs, Critical Thinking and Metaprogramming, Nanotechnology, Memetics, Space
Habitation, Spontaneous Orders, and Artificial Life.
        Extropy is now a full-size journal/magazine, typeset, and written by
computer scientists, philosophers, linguists, nanotechnologists, physicists
and others. No other publication is comparable in depth and range on topics
relating to technology and the future of the human condition.
.
.
Max More                more@usc.edu
Executive Director
Extropy Institute
.
...............................................................................
 ........
Yes, I would like to join the Extropy Institute and receive Extropy, Exponent,
other publications, and information on events.
.
USA:            $30*       Sustaining member:	$100
Canada/Mexico:  $33*       Corporate Sustaining
Overseas:       $40*            Member: 	$250
.
I would like to make a tax-deductible donation to help ExI with its projects:
.
o $10      o $25     o $50         o $100     o $500     o $1000     o Other: $
.
I don't want to join the Extropy Institute at this time, but do want to read
Extropy: The Journal of Transhumanist Thought. Enclosed is $9
($10 Canada/Mexico; $13 overseas)  for a year's (two-issue) subscription.
.
Make checks payable t o, and mail to: P.O. Box 57306, Los Angeles, CA
90057-0306.

Inquiries: (213) 484-6383. E-mail: more@usc.edu
.
* If you are already paid up to Extropy #11, deduct $9, or deduct $4.50 if you
have paid up to #10.
.

------------------------------

Date: Sat, 13 Jun 92 2:43:48 PDT
From: uunet!usc.edu!more@nunki.usc.edu (Max More)
To: Extropians@gnu.ai.mit.edu
Subject: Re: Ethics and objectivists

> It would be more accurate to say that the objectivists see that you have a
> fundamental choice, that of life or death, and that which you choose is
> not handed from above, but *if you choose life*, then blah blah blah.
>
> One may find this "fundamental choice" arbitrary, meaningless, or simply
> ill defined, but they derive a philosophy from a subjective choice between
> life and death.
>
> ......................
>
> Buy Buy -- Dan Davis
>

The Objectivist view of the choice to live (and to live as a human qua human)
is uncertain. Rand herself expressed it two quite different ways. In
"Philosophy: Who Needs It" she put it the way you definined: Each person must
decide between, life and death as fundamental values.

But, in other works, she apparently argues that you *must* choose life.
HEr argument for this is puzzling and unconvincing. She seems to think that
"a value is that which one acts to gain and/or keep". A value involves
acting to gain or keep something. Values are either instrumental or
ultimate. An instrumental value is one which you hold as a means to something
else. For example, you might value punctuality because of its typical
results. An ultimate value is one which is not sought for the sake of anything
further.

Rand's argument seems to be that there must be an ultimate value, because a
series of values/goals without an ultimate goal is impossible. According to
Rand, life is the ultimate goal. Only life can provide the stopping point --
the place where the "for-the-sake-ofs" termiante. She makes an obscure
argument that the concept of value genetically depends on the concept of
life. In so far as I've been able to make sense of this, it seems to me
to be wrong. While *value* is dependent on *life*, the *concept* of value
is not dependent on the *concept* of life.

I believe Rand also seriously errs in thinking that values form a hierarchy,
life being at the apex. I think values form a web or network. There is no
fundamental goal, only an interrelated web of goals, a web whose constituents
and weighting change over time.

Rand obscures this by talking of *life* indeterminately. By *life* you
might mean the collection of all your goals and values. Or you might mean
bare survival as a necessary condition for any action. But the latter
interpretation would make life a necessary condition (an *instrumental*
value) and not an ultimate value, while the former empties "life  as a value"
of any useful contents - any and all goals could then be included (including
seeking death).

For the record, I have agreed with everything Perry has posted on the topic of
morality. Instead of looking for objective or "natural laws", I look to
game theory, decision theory, and enlightened self-interest for guidance.
However (a topic that would take a lot of explaining) I don't believe
that morality is arbitrary -- i.e., there *can* be moral argument and
not merely moral assertion. This is because most of us do in fact share some
fairly fundmental preferences due to a shared evolutionary heritage and
broadly similar environment. If someone just doesn't share these deep values,
then argument ceases to be possible.

I devised and revised the Extropian Principles on this basis. I do not claim
that they are "objectively correct" or "objectively better" than other values.
They are merelyt intended to explicitly articulate a world-view shared by most
of us on this list, and an unknown number of others.

Max More		more@usc.edu
Executive Director
Extropy Institute

------------------------------

Date: Sat, 13 Jun 92 14:59:50 PDT
From: uunet!mprgate.mpr.ca!janzen (Martin Janzen)
To: Extropians@gnu.ai.mit.edu
Subject: Re: Ethics and objectivists

(Another rainy Vancouver day -- perfect for discussing philosophy! :-)

Max writes about Objectivist ethics:
>[...]
>Rand's argument seems to be that there must be an ultimate value, because a
>series of values/goals without an ultimate goal is impossible. According to
>Rand, life is the ultimate goal. Only life can provide the stopping point --
>the place where the "for-the-sake-ofs" terminate. She makes an obscure
>argument that the concept of value genetically depends on the concept of
>life. In so far as I've been able to make sense of this, it seems to me
>to be wrong. While *value* is dependent on *life*, the *concept* of value
>is not dependent on the *concept* of life.

Well, not exactly "genetically".  Let me see if I can clarify this a
bit.  When thinking about concepts, Rand would start asking what's often
called "Rand's Question": "What are the facts of reality that give rise
to this concept?"  In this case, we can start by perceiving the objects
around us, and noticing that some can be categorized as animate and others
as inanimate.

Inanimate objects simply *are*; they need not, and can not, take actions
to maintain their state.  The concept of "value" doesn't apply to these
objects; it's meaningless to talk about something being "good" or "bad"
for a rock.

Animate objects, on the other hand, exist in a state that is transitory,
and must act in order to maintain this state.  The concept of "value"
recognizes the fact that some actions are more likely than others to
accomplish this.  In this way, it *is* dependent on the concept of life.

                                * * *

As Max and Dan point out, this does assume that our animate object
*wants* to maintain this state.  Furthermore:
>The Objectivist view of the choice to live (and to live as a human qua human)
>is uncertain. Rand herself expressed it two quite different ways. In
>"Philosophy: Who Needs It" she put it the way you definined: Each person must
>decide between, life and death as fundamental values.

Minor nit: Given the way in which the concept of "value" is derived, I
think that while it is possible to choose death -- to choose not to act
in order to maintain one's life -- it's not correct to say that one can
choose "death as [a] fundamental value", since it is a state in which
"value" is meaningless.

>But, in other works, she apparently argues that you *must* choose life.

I'm not sure where she does so explicitly.  Her novels consistently
present as heroes the people who do choose life and who act successfully
to maintain and enhance it, and as villains the people who choose death
for themselves and those around them.  But to someone who chooses death,
the concepts of "hero", "villain", even "must" (because it implies
"...in order to..."), are meaningless.  Maybe to this person, Rand and
her philosophy simply have nothing to offer.

                                * * *

On another topic:
>I believe Rand also seriously errs in thinking that values form a hierarchy,
>life being at the apex. I think values form a web or network. There is no
>fundamental goal, only an interrelated web of goals, a web whose constituents
>and weighting change over time.

The idea that values form a strict hierarchy is certainly an over-
simplification; a web or network is a much better model.  I do think,
though, that she has a point in categorizing values.  For example, I
"act to gain and keep" abstractions like knowledge, integrity, and
productivity; material things like neat hardware and software toys; and
momentary pleasures like cappuccino -- clearly all different "levels" of
values, if not exactly a hierarchy.

                                * * *

>Rand obscures this by talking of *life* indeterminately. By *life* you
>might mean the collection of all your goals and values. Or you might mean
>bare survival as a necessary condition for any action. But the latter
>interpretation would make life a necessary condition (an *instrumental*
>value) and not an ultimate value, while the former empties "life  as a value"
>of any useful contents - any and all goals could then be included (including
>seeking death).

Yes, her usage of "life" is often confusing.  She's not referring to
"bare survival"; she wouldn't, for example, consider it moral for a
starving person to kill someone else in order to take their food.  She
does talk a lot about life as "man qua man", by which she means "the
life proper to a human being".

So what's "proper"?  Again, her approach is to look at reality; in this
case, at the fact that man's means of survival is the ability to reason
about the things he perceives, and then to act on that reasoning.
Therefore the "proper" life for a human is one in which he is able to
use his reason and to act on it.  If prevented from doing so -- say, if
he is enslaved -- he may take great risks in attempting to escape, or
even kill himself, rather than to subsist in an "improper" life.

(Aside: It follows that it's immoral for me to initiate the use of force,
because my doing so negates your ability to act according to your reason --
you have to "do what I say or else!", even if what I say is un-"reason"-able.
This is how Objectivists derive the Libertarians' "Non-Coercion Principle";
for better or worse, they criticize L'ism (or should I say, *some* L'ians?)
for omitting the steps that lead to the NCP and simply accepting it as
if it were an axiom.)

>For the record, I have agreed with everything Perry has posted on the topic
>of morality. Instead of looking for objective or "natural laws", I look to
>game theory, decision theory, and enlightened self-interest for guidance.
>However (a topic that would take a lot of explaining) I don't believe
>that morality is arbitrary -- i.e., there *can* be moral argument and
>not merely moral assertion. This is because most of us do in fact share some
>fairly fundmental preferences due to a shared evolutionary heritage and
>broadly similar environment.

Exactly!  Because of our "evolutionary heritage", our means of survival
are not our superior teeth and claws, or our ability to run faster than
predators, or to camouflage ourselves; it's our ability to reason about
our world, and to act accordingly.  This common ground allows us to make
generalizations about ethics, and to engage in moral argument with other
humans.

(Another aside: Note that this isn't a "universal" basis for ethics; it
applies only in the context of human beings.  The NCP doesn't apply to
tigers; their means of survival are different, and therefore so is the
proper life of a "tiger qua tiger"!)

(Yet another aside: Note implications for the so-called "animal rights"...)

>If someone just doesn't share these deep values,
>then argument ceases to be possible.

Yeah, really!  Happens all the time, especially on the Net...  I sometimes
find that I understand *why* I'm disagreeing with someone, and that it's
because of a fundamental disagreement on these philosophical issues -- but
that if the other person hasn't considered them explicitly, it's very
difficult to get down to the heart of the matter.

>I devised and revised the Extropian Principles on this basis. I do not claim
>that they are "objectively correct" or "objectively better" than other values.
>They are merely intended to explicitly articulate a world-view shared by most
>of us on this list, and an unknown number of others.

I hope I've made my own world-view more explicit, and that it helps you
with yours as well.

--
Martin Janzen
janzen@mprgate.mpr.ca (134.87.131.13)

[If you have Max's permission to repost the portions of his article that
I've quoted, you have my permission to repost my comments as well.]

------------------------------

Date: 13 Jun 92 19:53:01 EDT
From: Michael Clive Price <uunet!CompuServe.COM!100034.3077>
To: Extropians@gnu.ai.mit.edu
Subject: Re: Ethics

Hal>  By choosing an ethical principle, I increase the likelihood of
Hal>  others choosing that same principle (in the statistical, rather
Hal>  than the causal sense).  (Hofstadter had a column about this
Hal>  idea in Sci. Am. a few years ago.)

Hal, do you, or anyone else on the list, have a more precise reference to
this idea?  A while ago it occurred to me that this type of reasoning
should make us show compassion, especially to primitive creatures, on the
grounds that this would make it more likely to be treated compassionately
by any super-beings we encountered (eg if the universe is a simulation or
if we meet advanced aliens).  But I never found the idea treated
anywhere.

Same applies to copies.  If I've been copied then my "good" (however you
wish to define it) behaviour increases the chance of good behaviour in my
copies.  Presumably this "effect" is weakened the less the commonality
between the individuals involved.

Mike Price

------------------------------

Date: Sat, 13 Jun 92 20:24:54 PDT
From: uunet!ghsvax!hal (Hal Finney)
To: Extropians@gnu.ai.mit.edu
Subject: Ethics

   Hal>  By choosing an ethical principle, I increase the likelihood of
   Hal>  others choosing that same principle (in the statistical, rather
   Hal>  than the causal sense).  (Hofstadter had a column about this
   Hal>  idea in Sci. Am. a few years ago.)

Mike Price acts:

   Hal, do you, or anyone else on the list, have a more precise reference to
   this idea?

This article can be found in Hofstadter's book, "Metamagical Themas",
a collection of his Sci. Am. columns.  Chapter 30, "Dilemmas for
Superrational Thinkers, Leading Up to a Luring Lottery", was first
published in the June, 1983 Scientific American.

Hofstadter describes how he sent a letter to twenty of his friends
asking them to specify "Cooperate" or "Defect" in a prisoner's dilemma
game that he would run.  He would take each reply and provide a payoff
based on playing the game against each of the 19 other people.  The
matrix used was the classic:

			Cooperate	Defect
	Cooperate	(3,3)		(0,5)
	Defect		(5,0)		(1,1)

If everyone defected, they would each receive $19 (19 other players
times $1 each from the Defect,Defect entry); if everyone cooperated,
they would each receive $57 (19 times $3).  But Defecting always gets
your more money than Cooperating.

Hofstadter emphasized that each person was supposed to analyze the
problem rationally, realizing that everyone else would do the same
kind of analysis.  He hoped that most people would "therefore"
conclude that they should Cooperate, since if everyone did the same
thing they would all do better Cooperating than Defecting.

To his disappointment, it didn't work out that way.  14 people
Defected, and only 6 Cooperated.  However, I must say that the
analyses by his friends that he reported were remarkably shallow.

Dorothy Denning, for example, a well-known expert on computer
security, said, "I figure that if I defect, then I always do at least
as well as I would have if I had cooperated.  So I defect."  Well,
_duh_, Dorothy, to use an expression from my childhood.  For this you
went to college?  With this depth of reasoning the Prisoner's Dilemma
would never have been a problem at all, would it?  Just defect, that's
all, what's the big deal?

Even Martin Gardner moaned and groaned about how complicated the
problem is and ended up defecting.  Axelrod defected without
hesitation.  In fact, none of Hofstadter's friends were reported as
following the reasoning that was invited by the letter: figure that
what you did, your fellows were likely to do, too, and so Cooperate.

Hofstadter went on to pose a similar challenge to all readers of his
article, whom he estimated to number about 10,000.  You could enter
his contest by sending in a postcard telling how many times you wanted
to enter.  All such numbers of entries were added up to form the sum
N, and the prize would be given to a randomly chosen entry in the
amount of $1,000,000/N.  It turns out that rationally the best thing
to do is to enter once with probability about 1 in 7,000, else not to
enter at all.

I actually did this, when I read the article many years ago.  I
thought there would be more readers, so I estimated that I should
enter with probability 1 in 100,000.  My wife and I chose a five digit
number, opened a random number table from the appendix of an old
statistics book, and chose a random five digit number.  If they had
matched, we would have sent in our entry, but they didn't.

Needless to say, many people didn't do this, but instead tried to send
in huge numbers of entries so they would be sure to "win".  Hofstadter
seemed rather bitter and disappointed about this, and ended his
column-writing career in September, 1983 on this sad note.

I wonder what would happen if we were to have a similar contest here
on the list.  Does the Extropian philosophy of self-interest make us
more likely to Defect?  Or will our long-term viewpoint and common
attitudes give us the courage and hope to justify Cooperating?

I'm thinking of repeating Hofstadter's challenge for list members.  I
need to decide what stakes I can afford to offer in the no doubt
unlikely event that everyone Cooperates.  Stay tuned!

Hal Finney
ghsvax!hal@uunet.uu.net				...!uunet!ghsvax!hal
Permission is granted to freely reproduce this message.

------------------------------

Date: Sun, 14 Jun 92 3:58:23 EDT
From: uunet!gnu.ai.mit.edu!rjc
To: Extropians@gnu.ai.mit.edu
Subject: no subject (file transmission)


 re; Hal's prisoner dilemma post.

  I think that particular experiment is an inaccurate model for human
interaction because it lacks a time component and information transfer.

  In a real society, people would be able to get information on
who is a cheater and who is not (from past transactions). Over time,
people would tend not to interact with cheaters and tend to
associate with cooperators. In the end, cheaters will find that their
supply of suckers starts to dimish while the cooperators tend to dominate.

(cheater=defector)

  When there is only one interaction to be made, most people would tend
to defect. After all, if you're never going to interact with this person
again, why do you care if you cheat them? However, in the long term,
your reputation becomes important.  If you are going to interact with
the same people over and over, you would tend to cooperate with known
cooperators.

  I propose that a new experiment be conducted where the prisoner's
dilemma is conducted multiple times on the same group of people, and
full disclosure of the events is given. That is, you are told the outcomes
of all trades, who cheated, and who did not.

  Even if you are not told the outcome of everyone else's trade, as long
as you know the identity of people who cheated you in the past, you
could construct a trading strategy with avoids potential cheaters.

  -Ray

------------------------------

Date: Sun, 14 Jun 92 01:15:45 PDT
From: uunet!xanadu.com!tribble (E. Dean Tribble)
To: Extropians@gnu.ai.mit.edu
Subject: Ethics

I understand why Axelrod defected.  It is in fact in your interest to
defect in the Prisoner's Dilemna.  It is in your interest to cooperate
only in the Iterated Prisoner's Dilemna.  The iterated situation is
much diffent, because reputation matters.

dean

------------------------------

Date: Sun, 14 Jun 92 10:06:08 PDT
From: uunet!ghsvax!hal (Hal Finney)
To: Extropians@gnu.ai.mit.edu
Subject: Ethics & the prisoner's dilemma

   From: uunet!xanadu.com!tribble (E. Dean Tribble)

   I understand why Axelrod defected.  It is in fact in your interest to
   defect in the Prisoner's Dilemna.  It is in your interest to cooperate
   only in the Iterated Prisoner's Dilemna.  The iterated situation is
   much diffent, because reputation matters.

Well, you ought to read Hofstadter's column.  It is an extended
argument for why you should cooperate even in a single-shot Prisoner's
Dilemma.  The basic point is that whatever strategy you choose is
likely to be chosen by others.  If you choose to cooperate, it's more
likely that others will also do so.  This is because the situation you
face is the same as everyone else's, so whatever strategy reason
guides you to is the same strategy it will guide others to.

There are lots of circumstances where this kind of reasoning applies.
For example, should you vote?  I claim that if you don't vote, chances
are other people like you won't vote either, because you'll be the
kind of person who doesn't vote.  Hence your choice not to vote makes
it that much less likely that the candidate you would have voted for
will get votes.

OTOH, if you do vote, then you're the kind of person who votes, so
people who are like you will probably vote, too, and so whomever you
vote for will benefit not only from your vote, but from the votes of
people like you.

It's kind of a funny argument, I know, and relies heavily on the
(possibly weak) assumption that other people are much like you.  I
think Mike Price's example was better: suppose you were playing a
one-shot Prisoner's Dilemma against a recent copy of yourself.  This
example makes it even more likely that the other person will do the
same thing you decide to do.  Would you defect even in this situation?

Hal Finney
ghsvax!hal@uunet.uu.net				...!uunet!ghsvax!hal
Permission is granted to freely reproduce this message.

------------------------------

End of Hal's Extropians Digest Volume 92 Issue #0005
****************************************************