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

[Kevin Venzke]

Hi Mike,



> De : Michael Ossipoff <email9648742 at gmail.com>
> 
> On Tue, Oct 15, 2013 at 2:07 AM, Kevin Venzke <stepjak at yahoo.fr> wrote:
>>>  De : Michael Ossipoff <email9648742 at gmail.com>
>>>  But others here say or imply that it won't be a problem.
>> 
> Hi Kevin--
> 
> You wrote:
> 
>>  I hope I don't imply that. I think it's a major problem, likely to 
> be solved mostly by parties pressuring their clones > > against running in 
> the same election.
> 
> [/quote]
> 
> Yes, if those parties are mutually co-operative and amicable. But I
> don't think that that's the case. These groups aren't going to say,
> "Ok, you run a presidential candidate, and we won't, and we'll vote
> for your candidate."

I don't guess that internally parties are mutually cooperative and 
amicable, just that a consensus (among voters and commentators on 
strategy) will tend to find that it's too dangerous not to settle on 
"the" clone who's the "primary" nominee of the bloc.

Certainly, I don't think everyone will go along with this. It is not 
hard to imagine a Green and a Democrat, considered as representing a 
single large bloc, where neither candidate will drop out and neither 
loses all of their voting support to the other. But I don't think 
those candidates are likely to be clones either, not if you ask the 
voters. I expect plenty of Green supporters would deny that the 
Democrat is a second preference at all, and vice versa.

Suppose in that situation that between the Greens and Democrats
there is a pairwise majority over the lone Republican, but it's not
a mutual majority, due to a lack of preference sharing. In that case
an approach that wants to punish defections and elect the Republican,
has to assume that there was actually some deliberate defection in 
the voters' minds. If that's not so, then the outcome is being ruined
for no reason. (At least, I can't imagine we would be seeking to 
compel voters to list compromise choices that they don't actually 
consider to be such.)

It seems like for defection punishment to be safe, the candidates 
within a (potentially majority-strength) bloc, and the voters 
supporting them, would need to acknowledge, in some spirit of 
cooperation, that they are indeed a single bloc, and that the 
election is "(all of) us vs. them." Maybe you agree. It's just odd
to me that you don't fail to use terms like "mutually uncooperative"
to describe a progressive bloc.

> You continued:
> 
> I generally don't focus on elections with clones for this reason.
> [/quote]
> 
> But shouldn't these proposals, suggestions and discussions be about
> the acual existing situation?
> 
> The practical reality is that the progressives consist of many
> mutually un-cooperative and (in their perception) inimical parties and
> factions. But surely all progressives would rather have some
> progressive party in office, instead of the Republocrats--even though
> they have no agreement or cooperation regarding which one. Each party
> says, "We're the alternative. We're the answer."

Would this change if they managed to comprise a majority? I have 
trouble seeing why this situation is different from e.g. having 
several Democrats who want to run against a single Republican (say, 
an incumbent).

> 
> I admit that there almost surely isn't going to be any improvement or
> reform of any kind. There isn't reallly any hope for human affairs
> ever being any better than they are now (worse, sure). But,
> nevertheless, if discussing or proposing voting systems, it makes
> sense, to me, to discuss or propose them based on the
> almost-surely-false assumption that we're talking about a possible
> scenario. So then it makes sense to speak and propose in terms of the
> most nearly possible scenario, however impossible it might be.
> 
> People at EM tend to have a different approach to voting systems. They
> seem to regard it as more of a game, with no relation to the question
> of what could accomplish something,  if anything could. Since almost
> surely, nothing will be accomplished, it's only a game anyway. It's
> just that I'm playing a more realistic version of it than most EM'ers
> are.

It is a bit like a game, sure. I do think that the specific rules 
of the political system mostly determine the nature of the outcomes, 
meaning (if true) that it's safe and preferable to stick to 
considering election methods in an abstract context that can be 
studied with simulations, etc.

> 
> You continued:
> 
>>  You seem to be focused on a very specific scenario, where it is very easy 
> to tell (from the scenario definition) who's defecting from which camp and 
> how to punish them for it.
> [/quote]
> 
> Well, the trouble is that you can't know for sure that the other party
> will defect, or that they won't.

What I mean is, if you tell me that B and C are the Green bloc and the
B voters don't support C, I know that's defection, but only because you
told me that B and C are in the same bloc. Normally, elsewhere, I would
not have that information.


> 
> Anti-defecion strategy consists of partial defection. Forest suggested
> that anti-defection strategy a few years ago.
> 
> That partial defection, as an anti-defection strategy, would always be
> difficult guesswork. The guess would often be wrong. Often it would
> fail to prevent successful defection. Often it would give the election
> away to the Republocrats, even if both parties "co-operated" to the
> extent of trying for the optimal partial defection (which is also
> partial co-operation).
> 
> The strategy would only have deterrrent effectiveness if it created a
> strong probability of Republocrat win, in the event of the other
> party's defection. But, in order to ensure that, there would also be a
> significant probability of giving it away to the Repulocrats even if
> both parties did the partial cooperation/partial defection.
> 
> As you suggested, that problem is further compounded by the fact that
> there aren't just 2 players in that chicken-dilema game. There are
> lots of progressive parties, with many rivalries and animosities, at
> various different levels of the progessive family-tree.
> 
> That anti-defecin stategy, therefore, hardly qualifies as a solution,
> but it's all that we'd have, with a chicken-dilemma vulnerable voting
> system. Much better to just not use a chicken-dilemma-vulnerable
> voting system.
> 
> That's  why the progressives would need a chicken-dilemma-proof
> method, such as Benham or Woodall.
> 
> Sure, you might say that if we could elect a progressive with
> Plurality, by using, as a progresive-primary, the first presidential
> election in which progressive votes constitute a majority, then we're
> co-operative enough to not need a defection-proof voting-system.
> 
> But the whole point of better voting systems is to improve on
> Plurality, to reduce strategy-need. The progressives will want to be
> able to vote sincerely, without intra-progressive strategy, even if
> they can somehow overcome it to initially elect a progressive govt via
> Plurality.
> 

Sure. If they can enact a system that ensures they always have
incentive to vote as a mutual majority, then they don't have any
need of e.g. SDSC.

But this is a long ways off. Isn't there a sense in which it's
"more realistic," as you say above, to be concerned about whether
methods satisfy SDSC, or other criteria which could be useful to
parties that can't win in the short term but want to at least 
collect their share of the votes? I tend to think that there must
be many progressives who care about being able to support the 
"lesser evil" compromise, and who wouldn't be able to accept your 
advice to just vote sincerely, no matter what.

Kevin Venzke