[EM] Chicken Dilemma--To whom is it a problem? KV to MO

[Kevin Venzke]

Hi Mike,



----- Mail original -----
> De : Michael Ossipoff <email9648742 at gmail.com>
> 
> Kevin--
> 
> You wrote:
> 
>>  I would agree that an MMC violation would be a greater outrage than
>>  an SDSC violation. But I think MMC's premise would apply exceedingly
>>  rarely. That is why I think MMC is not a very useful guarantee.
> 
> Sure SDSC's premise would apply more frequently than that of MMC. That
> used to put me off, too, about MMC.
> 
> But if some set of voters agree on certain basic important things,
> differing only in details, and in their beliefs that their own party
> is the solution, then is it unlikely that theyd all prefer eachother's
> policy-proposals to the alternatives?
> 

Maybe I overstate the unlikelihood. The scenario doesn't make a lot 
of sense to me though. There is only one seat to win, yet all of 
these voters are reciprocating preferences for multiple candidates.
Whether it's because those candidates are really *that* similar and 
their voters are very cooperative, or whether instead the candidates
or parties are needing to actively plead with the voters to swap 
preferences, it seems like it has an inherent risk (with poor reward)
that would make it rare.

Bit of a tangent: The method I like for your scenario is a type of
open party list. Say each party can nominate as many candidates as
they want to run under that list. Then the voters can approve as
many lists as they like, and approve any candidates on any lists.
Then elect the most approved candidate on the most approved list.

To be effective maybe we have to ban approving candidates on the lists
you don't vote for. As a voter I certainly want that ability, but
it's possible that the threat of meddling from outside the party
would discourage parties from accepting a "moderate" on their list if
they think they have a change to win with a more "off-center" nominee.

Even so, I don't know if it would work. Whoever is crafting the party
list needs to be accepting of diversity among their candidates or
there isn't much point to it. Your progressive component parties
might end up having to run on their own lists anyway, turning the
method into plain Approval.



> 
> It seems to me that it's natural that if the A voters think that B's
> policies are similar enough to their own, so that they prefer B to C,
> then the B voters would also perceive that policy-similiarity. So
> mutuality doesn't seem like it would be rare.
> 
> Yes, I know that one scenario has B somewhere between A and C, maybe
> halfway between, resulting in indifference between A and C. But how
> likely is it that B would be exactly halfway inbetween? Isn't it much
> more likely that B would be closer to one than to the other? ...in
> which case there would be a {B,A} mutual majority, or a {B,C} mutual
> majority.
> 

Sure, but that's in sincere terms. If the sincere mutual majority is
{Bush,Gore} and those are the perceived frontrunners, then that (large)
coalition won't be visible on the ballots.

If the sincere MM is {Gore,Nader} but the perceived frontrunners are
Bush and Gore then I think you have an uphill battle convincing the
Gore voters to play nice. Nader voters are in the ugly position of 
having to ask the Gore camp for support. They are trying to sell the
story that they deserve to win if they have the votes. Yet they are
having to ask for votes to do it.

Under an SDSC method of course the Gore camp doesn't need to do
anything in this scenario. With a CD method it's less clear. But
without a change in the perceived status quo (of Gore being stronger)
the Nader voters can be conceivably blamed for threatening to wreck
an election over a candidate who has (without concessions from the
Gore camp) no chance of winning.

I think that the Gore camp can tell better stories and so will
probably truncate with clear conscience, and Nader voters will
probably see that, which reinforces still the likelihood that the
Gore camp will win the game of chicken.

If we have no idea which of Gore and Nader is the stronger, then
I have no idea what happens. I doubt this is a stable situation
that would survive for long. If it persists even up to election day
I guess this coalition will have major problems.


> 
> B voters being indifferent between A and C; and C voters being
> indifferent between A and B--That doesn't seem likely. 
>

I think that sums up the Bush/Gore/Nader scenario quite well though,
at least in terms of what the cast rankings would probably look like.

>> 
>>> 
>>>  I don't know if MMC,CD, CC and SDSC are compatible. Probably not,
>>>  because I don't know of a method that meets them all.
>>> 
>> 
> My immediate question would be whether
>>  CD is sufficiently distant from LNHarm that it could be
>>  compatible with Condorcet
> 
> CD is compatible with Condorcet. Benahm and Woodall comply with both CD and CC.
> 
> They both comply with CD and Smith.
> 

Did you say CD will be limited to the three-candidate case? I do 
wonder what happens when the "enemy" is actually a coalition with
more candidates than the progressives. I think that would make it
hard to define a method which, when the voted mutual majority is
broken, absolutely never picks the candidate who isn't supposed to
win.

Kevin Venzke