[EM] burden of doubt and IRV/Condorcet politics

[James Green-Armytage]

James replying to Mike...

>It isn't clear what you point is. 

	I've already clarified it somewhat in a previous post. "The burden of
doubt should be cast on any new voting method, and that voting methods
designers should actively hunt for the flaws in theoretical systems rather
than assuming those flaws to be nonexistent until they occur in practice." 
	Actually, I don't really even need to say this. Whenever a new voting
method is seriously considered, its advocates *do* shoulder the burden of
doubt in practice. Hence, I'm merely saying that we should prepare
ourselves to do this for any method that we would like to implement. If we
can't anticipate and effectively answer the worst criticisms that can be
made of our favorite systems, then we will be unprepared when someone else
brings them up at a critical juncture.
>
>Sometimes CVD people complain that 
>Condorcetists and Approvalists are working against reform when they
>oppose 
>IRV proposals. 

	I currently agree with this perspective.

>But it was Condorcetists and Approvalists who asked Richie to 
>avoid this situation in which IRV's faults are discussed publicly.

	Nothing wrong with an honest public discussion of method pros and cons.
But the point remains that IRV is an improvement over plurality and
runoffs, and hence opposing IRV amounts to opposing an improvement.
	I do find the relative close-mindedness of some IRV advocates to be
frustrating at times. However, I don't think that there is much if
anything to be gained by responding with hostility, and trying to block
mildly good proposals because their advocates are not able or willing to
appreciate better proposals.
>
>I've never heard of an IRV proposal succeeding where there was any 
>opposition from Condorcetists &/or Approvalists. IRV failed in Alaska,
>and 
>initially in San Francisco. On both occasions, a few of us heard of the 
>proposal, and sent in our comments. Condorcetists and Approvalists can
>and 
>will sink IRV anywhere where they have the opportunity and the time to
>tell 
>the people what IRV is like.

	I hope not. Aside from yourself, who exactly are these anti-IRV
pro-Condorcet people? It seems to me that most people who understand the
merits of Condorcet methods should appreciate that IRV is at least a small
step in the right direction. Who else, besides yourself, has actively
opposed (lobbied against) IRV as a replacement for plurality and runoffs
on the grounds that it is inferior to Condorcet or approval?
	Your paragraph above seems somewhat self-serving, in that you are taking
credit for IRV adoption failures where there were obviously many other
factors at work, and making rather vague excuses for the cases that don't
fit your grandiose assertion.
	In general, I hope that you are acting fully in the interest of a better
democracy, rather than compromising this pursuit to the service of your
own ego. Obviously it stings to be ignored and marginalized by the
mainstream IRV advocates, and hostility is an instinctive response, but
that doesn't mean that it is the most productive response in the long run.
>
>Really, we should have an organization dedicated to finding out about
>each 
>IRV proposal in the U.S., or anywhere, and taking turns writing to the 
>decisionmakers involved in the choice, or getting our information
>published 
>in newspaper letters or ballot-pamphlets there, attending forums, etc. We 
>could fairly divide the time and labor of doing that work. I can't do it 
>alone.

	The question is, who else would want to devote their time to a group with
such a negative goal? I know that I wouldn't. I think that an organized
pro-Condorcet group would be a great idea, but I'd much rather promote the
methods I like than waste my effort trying to oppose IRV, something that I
consider to be at least a small step in the right direction.
	Furthermore, if Condorcet fans actively oppose IRV, IRV fans are very
unlikely to provide assistance when a Condorcet group tries to get
Condorcet adopted somewhere. Your strategy of opposition (assuming that it
is effective, which is questionable) seems to lead to a no-win situation.
	Later in my life, I might attempt to co-found a national democracy group
that emphasizes more advanced systems like Condorcet, STV, and delegable
proxy, but for the time being I have no definite plans in that direction,
as I expect to devote the next several years to academic study.
>
>But, James, to return to the original topic, IRV is already being 
>aggressively promoted around the country. Do you seriously suggest that
>we 
>should discuss at great length how "badly" wv can fail (in ways that I 
>consider relatively irrelevant because they aren't strategy problems)? 

	Yes, I do seriously suggest that. This is obviously worth knowing if we
intend WV for use in large public elections, and it is clearly germane to
the focus of this list.
	Furthermore, if WV is being seriously considered for use in a large
electorate, this issue *will* come up. I will not have to be the one to
bring it up. Most people on this list are relatively casual about
strategic manipulation in Condorcet methods, but many IRV people take a
much more dire view on the matter, and they will make their voices heard,
especially if you persist in trying to undermine their pro-IRV work. They
can rather easily make a case that margins is too strategically unstable
for public use, and then when you/we say that WV reduces that problem to a
negligible level, they will quite rightly demand a rigorous argument from
you/us, to support your/our assertion. 
	So, if WV does reduce the strategy problem to a negligible/acceptable
level, then we might as well get started on our rigorous argument to
support this. If it doesn't, then I might not want to support it. Either
way, the discussion is important.

>I 
>suggest that when IRV is being gung-ho promoted by people who don't care 
>about its problems, this is no time to waste our time debating wv
>endlessly 
>before offering it as an alternative to IRV. 

	You or anyone else can offer it as a public proposal whenever you like.
If you're confident in WV, you should lobby for it yourself rather than
only urging other people to do so. As for me, I am not yet fully convinced
that WV can provide stability. Maybe it can, maybe it can't; I have no
conclusion to offer at this time. 

>Yes, as I said, I consider CR 
>to be a better public proposal, but some suggest that a multi-pronged 
>approach is good too. 

	Of course. Why not? Lobby for CR if you like CR. Lobby for WV if you like
WV. Lobby for both if you like both. They're both interesting methods...
let's see what happens.

>I've heard from a lot of people who like wv, including 
>a state lawmaker who has proposed IRV in his legislature, and was
>interested 
>in proposing wv there.

	Great. What's his name, and what district does he represent? Condorcet is
surely a great idea, no question about that. In the long run, if the world
tends toward greater democracy, I believe that Condorcet will eventually
be embraced for wide use. 
>
>So, James, since it's the IRVists, not we, who are aggressively proposing 
>something that they don't understand the problems of, maybe it would make
>a 
>lot more sense for you to be discussing the need to talk about IRV's 
>problems instead.
>
	I think that IRV's problems are already well-understood on this list,
e.g. Condorcet failure, Smith failure, relatively frequent compromising
incentive. Hence, I don't feel a strong need to discuss those problems on
this list. I do discuss them with IRV advocates, although I try to do so
in a productive and non-alienating manner. I've also been trying to
convince some IRV advocates that equal rankings should be allowed in IRV.
I think that IRV's more significant problems (not participation and
consistency, I know, but we agree that those are unimportant) are shared
by plurality and runoffs, and so I don't consider them sufficient grounds
to oppose IRV as an alternative to those two methods.

Sincerely,
James
http://fc.antioch.edu/~james_green-armytage/voting.htm