On Aug 15, 2011, at 10:31 AM, Jameson Quinn wrote:
> 2011/8/15 Jonathan Lundell <jlund...@pobox.com>
> On Aug 15, 2011, at 8:35 AM, Jameson Quinn wrote:
>
>> 2011/8/15 Jonathan Lundell <jlund...@pobox.com>
>> On Aug 15, 2011, at 7:09 AM, Jameson Quinn wrote:
>>
>>> So, what do you think? Let the debate begin. I expect the above to be torn
>>> to shreds. But once it's starting to seem stable, I'll make a google doc
>>> out of it, so we can collaboratively polish up the language.
>>>
>>
>> Where you will lose many of us, I think, is in a flat endorsement of
>> approval. The fundamental problem with approval voting is that, with more
>> than two candidates, voting demands that the voter engage in strategic
>> voting. That is, if my preference is A>B>C, then my decision whether to
>> approve B cannot be made without strategizing. That flies in the face of
>> your fine suggestion that strategy avoidance be an important criterion.
>>
>> (It's not an answer to say that approval strategy is "easy" or "obvious";
>> that's not the point, nor is it generally true, since it depends on having
>> information not generally available about other voters' preferences and
>> strategies.)
>>
>> Well, I specifically didn't make such a claim, because, although I believe
>> it, I knew it would lose people.
>>
>> I would claim that approval strategy is at least as easy/obvious as
>> plurality strategy; and that approval reacts at least as well to a lack of
>> strategy as plurality.
>>
>> The problems of IRV are minor compared to approval (and any other
>> rating-based system).
>>
>> You can believe that and still sign this statement, as long as you believe
>> that approval is a worthwhile step up from plurality. Note that the
>> statement nowhere claims that approval is better than IRV, just that it is
>> likely to be able to get a broader consensus from theorists.
>>
>> Still... although you haven't actually said anything that contradicts what I
>> wrote, you did say that I will lose you for endorsing approval, which makes
>> me suspect that you also feel that approval is not a worthwhile step up from
>> plurality. I'd love to convince you otherwise, but the most important thing
>> is to get a worthwhile consensus statement. To me, any statement that can't
>> flatly endorse even one system is meaningless, and I'd guess approval is
>> probably the system which can get the broadest support, and also one of the
>> few which has a real chance of being implemented for real-world political
>> elections. Am I wrong? Would you, for instance, endorse SODA?
>>
>> Personally, I see voting reform as a step-by-step process. Yes, approval has
>> unsatisfying aspects. But implementing approval is a clear step up from
>> plurality; a clear step towards any system you might advocate; and a step
>> that would give us useful empirical data to help decide which direction to
>> go from there. I think that most voters would be wary of taking a larger
>> leap, even to my one of my favored systems, which I think are among the
>> simpler of the better options. And since the simpler, safer option of
>> approval does exist, I can't even blame voters for that.
>>
>> So, the bottom line is: Jonathan, what do you suggest? Do you think that
>> this statement would still be useful if we simply removed approval and thus
>> made no clear policy suggestion? Do you think that there's some other system
>> which could get broader support from this community than approval would? Or
>> do you thing that a useful, broadly-supported statement is simply
>> impossible? (Or do you see some other option which I don't?)
>>
>
> I'm doubtful that such a statement is possible, at least if you want
> universal assent.
>
> Not universal, just as broad as possible.
>
> If we grant for the sake of the argument that approval is strictly preferable
> to plurality (and I might argue with that), then I don't see the case for
> rejecting IRV out of hand. Sure, there are list members who are allergic to
> IRV, but it seems to me that the argument against strategy-intensive rules is
> at least as convincing (more so to me).
>
> There are people here who have reasonable arguments that IRV is strictly
> worse than plurality, and so will not sign any statement advocating it. You
> are making a reasonable argument that to me says that Approval is only
> marginally better than plurality. So I don't see how that precludes you from
> being able to sign on to this statement.
>
> Perhaps your argument is better than the anti-IRV argument, but that's not
> the point here. We can certainly assume that it's impossible that everyone
> here will agree on any statement of that form.
>
>
> In general, if the primary reason we're rejecting plurality is its strategy
> problems (and your three listed problems, honest results,
> strategy-resistance, and candidate incentives all fall into a strategy
> category, seems to me), then it's hard for me to see a strong motivation for
> advocating approval.
>
> Again, broadest possible consensus. Approval is nobody's favorite (I think;
> it certainly isn't mine), but it's a good next step, and something we can (I
> hope) agree on.
>
It's true that I might agree to a statement if all it said were "We believe
that approval is marginally superior to plurality" (thought to the extent that
I agreed, I don't think it's enough better to merit any energy in advocating
it). But that's not what you're proposing. Is it?
----
Election-Methods mailing list - see http://electorama.com/em for list info
