[EM] How to fix the flawed "Nash equilibrium" concept for voting-theory purposes

Abd ul-Rahman Lomax abd at lomaxdesign.com
Sat Apr 24 20:36:47 PDT 2010


At 06:56 PM 4/23/2010, Warren Smith wrote:
>And, you may not have noticed what I said re humans were designed by
>Darwin for smaller group sizes, e.g. tribes of a few 100 members, and
>their notions of "rational" are designed for groups of those sizes.  I
>think a lot of behavior about sizes larger than that (such as a
>country-wide election) can be understood roughly, by saying "humans do
>stuff that'd be rational if it were size<300.   The human inbuilt
>pseudo-rationality
>device basically can only count up to 200 and all populations>200 are
>treated by it as 200."
>
>Mind you, this is just my speculation.  Can anybody see a way to prove
>it? I can't.

I don't know about proof, but I do think that we are "designed" for 
small-group decision-making. And thus my tendency is toward designing 
systems that work on a small scale and then arranging structure so 
that decisions are amalgamated through a hierarchy of "meetings" of 
some kind, with each "meeting" being relatively small scale, the 
people participating in it know and directly interact with each 
other. And there is a lot of back-and-forth, not just a flow of 
decision-making toward the center. That's what delegable proxy should 
do, and Asset Voting would probably get there, too. People want 
palpable influence and effect. It's what is largely missing on a 
large scale that is present in small New England towns that have 
direct democracy.

Amherst has a huge Town Meeting that isn't. It's a representative 
assembly that is election on a very small district within the town, 
several hundred of them. Bad idea. The meeting is famous for highly 
contentious and drawn-out process, and it narrowly missed being 
abolished in two recent votes. Not the way to do it!

Asset would be beautiful for Amherst, in fact. Elect a manageable 
size of assembly, could be thirty, that would provide a great deal of 
diversity in representation, and the electors could still vote on 
issues if they wanted. (But not address the assembly or enter 
motions.) (Elector voting, I'd predict, would only rarely make a 
difference, and a roll call would have to be done for it, so that all 
votes were specifically recorded.)




More information about the Election-Methods mailing list