[psychohistory] What can a Society Do?

[Jim Choate]

On Wed, 12 Sep 2001, Chen Yixiong, Eric wrote:

> > Actually it isn't Godel's (which just says some statements can't be found
> > definitively true or false - it is undecidable). However, Arrow's
> > Impossibility Theorem does(!) do exactly what you want.
> 
> 
> According to the Theorem page at:
> http://www.personal.psu.edu/staff/m/j/mjd1/arrowimpossibilitytheorem.htm
> 
> I think I did not draw parallels to my writings below.

???? That sentence is without meaning as worded.

> The theorm seems to apply for democratic systems, but here I write about 
> systems in general.

But that's exactly(!) why Arrow's applies. Fairness (ie democratic) as
individual choice (ie vote).

The fact that there are no democratic (ie fair) mechanisms by which to
make selections is what damns Crypt-Anarchic-Capitalist-Libertarian (CACL)
thought. A good example is to apply voting theory to David Friedmans law
by contract examples.

One can cast CACL thought as democratic thought except for a single
difference. Democratic thought requires all to be responsible to the same
set of rules and policies. CACL doesn't.

CACL thought has three fundamental flaws. One, it requires individuals to
behave in a manner inconsistent with their biology. People ain't rational
in any sort of consistent manner. Second, it says that if we get rid of
government people will resolve their own problems. However, this statement
can be turned into a 'democratic election' (ie the participants have their
own choices to make - a vote) yet we can prove mathematically that such a
solution is impossible if there are more than two (2) participants. Now I
don't know about you but anything worthy of the label 'society' must have
more than two (2) people involved. Finaly, it assumes that all issues and
decisions one makes in life can be reduced to 'economic' decisions. While
that is certainly possible it is clear that not all issues are deal with
equitably in such a manner.

(It's also applicable to Tim May's commentary a while back about the
'solved n-division' problem (which can also be shown to be a vote if the
remainder is indivisible or can't be re-connected) and the fact that he
seems to think that the solution that is out there now is universal. When
in fact it applies to such a small fraction of n-division problems as to
be worthless. The proof is in the 'remainder' and it's characteristics.
Most of reality doesn't comply with the requirements for 'fair
divisibility' - too granular.)

> I think we had referred to different versions of Godel's Theorem, where
> I use this version : "No system of rules can have both completeness and
> consistency (including social systems applied to humans)".

But a social system isn't (just) a 'system of rules'. It also has beliefs
and various cause-effect and dependency issues related to environment and
biology that cause Godel's to be inapplicable. The fact that it consists
of anything(!) more than a system of rules is enough to invalidate
Godel's. Godel's simply isn't applicable to a broad enough set of
examples. Though Godel's may keep you from proving it). The reality is
that the concept of 'social system' is entirely too broad for the conept
of 'self-consistent language' to be applied. Where did the requirement for
'consistency' in a social setting come from in the first place? And what
does 'consistency' actually mean in that context? A naive interpretation
might be that they always make decisions the same way or perhaps the same
selection. Either will fail because it won't respond to changes in the
environment. Clearly in conflict with the premise of being 'consistent'.

 --
    ____________________________________________________________________

            natsugusa ya...tsuwamonodomo ga...yume no ato
            summer grass...those mighty warriors'...dream-tracks

                                            Matsuo Basho

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage at ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------