Re: [EM] majority tyranny (was proof idea for non-summability of STV)
An insight of Jobst Heitzig has made a big difference in my thinking about
this topic ... namely that all of the supposedly democratic deterministic
consensus-building techniques must (in order to guarantee reaching a full
formal consensus compromise in every case) sometimes resort to some kind of
more or less subtle group pressuring of some participants.
Once this insight gets lodged in the brain of someone who is categorically
opposed to coercion, compulsion, or abuse of conscience in any form, in any
degree, such a person must wonder if there exist non-deterministic,
abuse-free methods for achieving consensus.
To Jobst we also owe the insight that in the context of "lottery methods" a
non-coercive, proportionately fair, full consensus solution always does
exist!
Jobst has led the search for the simplest minimal entropy solutions. A
zero-entropy solution is necessarily minimal. If there is no zero entropy
consensus, then by definition there is no deterministic consensus.
There is a lot of psychological pushback against this idea ... which is
understandable ... it's the same psychology that made Einstein (and almost
everybody else) so reluctant to welcome quantum mechanics into mainstream
physics a century ago.
This psychology is perhaps reflected in the traditional designation "random
dictator" for the benchmark lottery more neutrally known as, "random ballot
choice."
Game theorists have known for ages that for some games no "pure" strategy
is optimal ... mixed strategies are sometimes indispensable. What is a
mixed strategy? It is a mixture of pure strategies. How are they mixed
together? Stochastically, i.e. by lotteries.
I'm not saying that every election methods scientist needs to specialize in
lotteries. I don't know as much as I should about them, but I'm glad that
Jobst has raised my consciousness about the limitations of determinism and
about the possibilities of minimal entropy excursions away from determinism
in the inerest of efforts towards social consensus as opposed to mere
"majority rule."
I hope nobody gets the impression that I am a know-it-all on this listserv;
I have huge gaps in my knowledge ... but the few things I think I
understand ... I try to articulate them as clearly as possible ...to
facilitate communication, including facilitation of revealing my mistakes
... which happens more and more often as I get older🙄
Date: Thu, 3 Dec 2020 18:04:40 +0000 (UTC)
From: Kevin Venzke stepjak@yahoo.fr
To: Richard Lung voting@ukscientists.com, EM
election-methods@lists.electorama.com, Kristofer Munsterhjelm
km_elmet@t-online.de
Subject: Re: [EM] Proof idea that IRV can't be summable
Message-ID: 1307822904.7547807.1607018681005@mail.yahoo.com
Content-Type: text/plain; charset=UTF-8
Le jeudi 3 d?cembre 2020 ? 04:20:54 UTC?6, Kristofer Munsterhjelm <
km_elmet@t-online.de> a ?crit :
If told to create something democratic without concern to current
constraints, I'd probably just skip right to sortition. This would not
just invalidate single-winner methods, but voting altogether; except,
possibly, the method the assembly itself uses to decide.
In this exchange "democratic" must mean that the assembly's seats are
allocated proportionally. This leaves the issue of allocation of actual
policy-making power as you suggest.
Maybe there is a way to determine policy proportionally, and without using
randomness. I don't think it can be based on decay of individual delegates'
voting power (because if you use your power sub-optimally you may fail to
influence policy) or on how many things you vote on (because proposals
could be clones of each other etc.). So it might have to be based on
time... A faction gets an amount of time in power.
But realistically there is probably a minimum faction size you would want
to allow to wield power. And to prevent whiplash you'd probably want a
minimum amount of time that a faction could be in power. Could a faction
representing 25% of the voters be allowed to set policy for even a year? If
not, can we defend that without invoking the principle of majority rule? (I
doubt it... And for me that is always the limitation, that no matter what,
you have to implement majority rule somewhere in the process.)
Kevin
On 12/03/2020 7:57 PM Forest Simmons fsimmons@pcc.edu wrote:
An insight of Jobst Heitzig has made a big difference in my thinking about this topic ... namely that all of the supposedly democratic deterministic consensus-building techniques must (in order to guarantee reaching a full formal consensus compromise in every case) sometimes resort to some kind of more or less subtle group pressuring of some participants.
Once this insight gets lodged in the brain of someone who is categorically opposed to coercion, compulsion, or abuse of conscience in any form, in any degree, such a person must wonder if there exist non-deterministic, abuse-free methods for achieving consensus.
To Jobst we also owe the insight that in the context of "lottery methods" a non-coercive, proportionately fair, full consensus solution always does exist!
but it's not repeatable, if it's truly a random lottery selection. if you were to base any random numbers (to make a decision) on a PRNG (that's repeatable), then it's not really random. you will know in advance which candidate is getting picked.
...
I'm not saying that every election methods scientist needs to specialize in lotteries.
But without repeatability, there is no recount. The only way an election can be recounted (if the law or courts required it) is if the RNG is a PRNG beginning with the same seed.
Kevin Venzke (i think) wrote:
But realistically there is probably a minimum faction size you would want to allow to wield power. And to prevent whiplash you'd probably want a minimum amount of time that a faction could be in power. Could a faction representing 25% of the voters be allowed to set policy for even a year? If not, can we defend that without invoking the principle of majority rule? (I doubt it... And for me that is always the limitation, that no matter what, you have to implement majority rule somewhere in the process.)
Nothing wrong with the principle of majority rule. Elections are about majorities. That's why Hare RCV failed in Burlington in 2009 and a Condorcet method would have complied with majority rule.
Drawing lots should only happen when, after recounting and litigating and scrutinizing, there is a dead tie.
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
On 04/12/2020 02.25, robert bristow-johnson wrote:
On 12/03/2020 7:57 PM Forest Simmons fsimmons@pcc.edu wrote:
An insight of Jobst Heitzig has made a big difference in my
thinking about this topic ... namely that all of the supposedly
democratic deterministic consensus-building techniques must (in
order to guarantee reaching a full formal consensus compromise in
every case) sometimes resort to some kind of more or less subtle
group pressuring of some participants.
Once this insight gets lodged in the brain of someone who is
categorically opposed to coercion, compulsion, or abuse of
conscience in any form, in any degree, such a person must wonder if
there exist non-deterministic, abuse-free methods for achieving
consensus.
To Jobst we also owe the insight that in the context of "lottery
methods" a non-coercive, proportionately fair, full consensus solution
always does exist!
but it's not repeatable, if it's truly a random lottery selection.
if you were to base any random numbers (to make a decision) on a
PRNG (that's repeatable), then it's not really random. you will know
in advance which candidate is getting picked.
You can make a random election repeatable by first recording the
entropy, then doing the count, then once everybody is satisfied that the
deterministic numbers are correct, only then reveal the entropy (and
seed a PRNG with it or something).
You can even get this entropy without explicit use of randomness by
cryptographic means. E.g. ask a group of people to submit a number of
(arbitrary) quotes, record them in sealed envelopes, then once the
deterministic variables are set, hash the quotes together and use the
hash as a PRNG seed.
The only difficulty relies in keeping the entropy secret until the count
is done.
Alternatively, you could allow for recounts after the entropy has been
unveiled, but then the recount has to be scrutinized more closely;
otherwise the different factions could try to only do recounts where it
would change the (now determinized) decision procedure in their favor.
I'm not sure any reference needs to be made to entropy. If you're using a non-deterministic election method, you do the ballot count as normal, and this gives us the probabilities for each candidate in the lottery. You can pause at this point and have a recount or multiple recounts. Then once the lottery probabilities are set, you proceed to do the lottery. Obviously this bit will only be done once, but many countries have their e.g. national lotteries which are held to a great degree of scrutiny, and normally involve numbered balls coming out of a machine. So something similar could be done here. Just be careful in South Africa! https://www.bbc.co.uk/news/world-africa-55154525
Toby
On Monday, 7 December 2020, 20:52:21 GMT, Kristofer Munsterhjelm km_elmet@t-online.de wrote:
On 04/12/2020 02.25, robert bristow-johnson wrote:
On 12/03/2020 7:57 PM Forest Simmons fsimmons@pcc.edu wrote:
An insight of Jobst Heitzig has made a big difference in my
thinking about this topic ... namely that all of the supposedly
democratic deterministic consensus-building techniques must (in
order to guarantee reaching a full formal consensus compromise in
every case) sometimes resort to some kind of more or less subtle
group pressuring of some participants.
Once this insight gets lodged in the brain of someone who is
categorically opposed to coercion, compulsion, or abuse of
conscience in any form, in any degree, such a person must wonder if
there exist non-deterministic, abuse-free methods for achieving
consensus.
To Jobst we also owe the insight that in the context of "lottery
methods" a non-coercive, proportionately fair, full consensus solution
always does exist!
but it's not repeatable, if it's truly a random lottery selection.
if you were to base any random numbers (to make a decision) on a
PRNG (that's repeatable), then it's not really random. you will know
in advance which candidate is getting picked.
You can make a random election repeatable by first recording the
entropy, then doing the count, then once everybody is satisfied that the
deterministic numbers are correct, only then reveal the entropy (and
seed a PRNG with it or something).
You can even get this entropy without explicit use of randomness by
cryptographic means. E.g. ask a group of people to submit a number of
(arbitrary) quotes, record them in sealed envelopes, then once the
deterministic variables are set, hash the quotes together and use the
hash as a PRNG seed.
The only difficulty relies in keeping the entropy secret until the count
is done.
Alternatively, you could allow for recounts after the entropy has been
unveiled, but then the recount has to be scrutinized more closely;
otherwise the different factions could try to only do recounts where it
would change the (now determinized) decision procedure in their favor.
Election-Methods mailing list - see https://electorama.com/em for list info