[EM] publicly acceptable election methods

[James Green-Armytage]

Hi Russ,
	It's good to see you so engaged in the list of late. I regret that I
haven't been participating much in the last few months.

>Let me just 
>suggest, however, that it would be useful to distinguish between 
>election methods that could be publicly acceptable within our lifetimes 
>and those that almost certainly will not be.

	I believe that such distinctions are regularly made here on the EM list.
I think most methods exist on a spectrum from the extremely practical to
the extremely theoretical.
>
>I believe that Approval is simple enough to be publicly acceptable. 

	Certainly.

>IRV 
>may or may not be. 

	The public acceptability of IRV has been demonstrated at various times
and in various places. IRV is now used in San Francisco, it is on the
books in Berkeley (where I currently live), and I believe that it was
recently accepted for use in Burlington, VT. In all these cases, IRV has
won public referenda. Hence, it is publicly acceptable to the people of
these cities. Of course, the cities are unusually liberal. Is IRV publicly
acceptable in middle America? That remains to be seen. IRV has also proven
to be publicly acceptable in other countries, e.g. Australia, Ireland,
etc. 
	The concept of IRV is also becoming more mainstream in political circles.
Note that Howard Dean (currently the head of the Democratic party, hence
clearly a mainstream politician) and John McCain have both expressed
support for the idea.
	
>As I have pointed out before, IRV will have serious 
>implementation problems due to lack of "summability," 

	I disagree that this problem is serious.

>Unfortunately, the same cannot be said for Condorcet as it is currently 
>understood. Imagine how befuddled the public will be about "dropping 
>defeats" (as opposed to the much simpler idea of dropping candidates). 

	Yes, this could be difficult. If you feel that this is the major barrier
to Condorcet methods being explainable, there is always the Raynaud
method. I argue that this is simpler than the method you have proposed.
http://userfs.cec.wustl.edu/~rhl1/rbvote/desc.html
	A brief definition: Eliminate the candidate with the strongest defeat
against him or her, until no defeated candidates remain. (Note that when a
given candidate is eliminated, his victories over other candidates (if he
has any) are removed from consideration.)
>
>Do we use Schulze, Ranked Pairs, or some other variation? 

	While I believe that beatpath and ranked pairs are equally good methods,
I suggest that ranked pairs is slightly more easy to explain.

>Then there's 
>the "margins vs. winning votes" issue

	It's tricky, I agree. My personal opinion is that winning votes methods
are preferable, though not perfect. My own pet method is cardinal
pairwise, although I realize that it does not do so well from a
simplicity/mass acceptability standpoint. I wonder if a Condorcet method
which provides a candidate withdrawal option would be publicly acceptable
in the near future... As you say, these issues make Condorcet adoption
more difficult.
>
>
>I've also been reading a lot here about "lottery" methods. I have the 
>utmost respect for those who are discussing them, most of whom are 
>probably smarter than me, but I must confess that I don't have a clue 
>about what they are trying to accomplish. I think it is safe to say, 
>nevertheless, that any method that involves a "lottery" is publicly 
>unacceptable. 

	I agree that lottery election methods will not be acceptable on a large
scale. I assume that the interest in the methods are mostly
theoretical/academic/mathematical rather than political/practical. In the
former capacity, I can certainly appreciate the interest involved,
although I admit that I too have been unable to keep up with that
particular thread.
>
>The voters rank the candidates and also specify an Approval cutoff. The 
>CW wins if one exists, otherwise the least approved candidate is 
>repeatedly eliminated until a CW is obtained.
>
>That's it, folks! Two sentences. Now, *that* could be simple enough to 
>be publicly acceptable. No complicated algorithms understandable only by 
>Ph.D.s are needed. 

	Personally, I disagree that this proposal is significantly more simple
than ranked pairs. I think that it is decidedly more complex than minimax
(the method Mike refers to as "plain Condorcet") and Raynaud. You may have
defined it in two sentences, but many terms in that definition will
require additional definition for the average layperson.
	A ranked pairs definition in two sentences: Consider pairwise defeats in
descending order of strength. Drop a defeat if it forms a cycle with an
(undropped) stronger defeat.
>
>I would really like to know what the exceptional folks here on EM think 
>about this proposal. Some of you have provided some feedback already, 
>and I certainly appreciate it. However, I don't sense any excitement 
>about it, and I'd like to know why not. What is the problem with it?

1. It combines two rather disparate principles (the pairwise principle and
the approval principle) in a way that I personally find inelegant
(semi-arbitrary). I prefer methods that have a kind of conceptual unity,
as opposed to methods that seem like an arbitrary combination of rules
with separate logical roots.
2. It is not Smith-efficient; that is, it doesn't always choose a member
of the minimal dominant set. I think that Condorcet methods that lack
Smith efficiency are somewhat inelegant. Definition of Smith set:
http://lists.electorama.com/pipermail/election-methods-electorama.com/2005-February/014742.html
3. The possibilities for tactical voting may well prove to be complex.
4. If you are primarily interested in simplicity, I think this method
suffers from the fact that it uses two distinct ballot types (binary and
ordinal). Of course, cardinal pairwise (as I usually describe it) uses
both cardinal and ordinal ballot types, although it can conceivably be
done with only a cardinal ballot. However, I don't tout cardinal pairwise
as being an extremely simple and salable method, as you are doing with
this method.

>If this proposal is seriously flawed, then I seriously doubt that 
>Condorcet has a chance of ever being used in large-scale public 
>elections in my lifetime.
>
	I would argue that the viability of pairwise methods in general is not
dependent on the viability of this particular proposal. Again, I suggest
that you take a look at the Raynaud method if you're worried about the
defeat-dropping concept. As to whether Condorcet will be used in your
lifetime, that might depend on how long you live. It will be interesting
to see which country, if any, will be the first to implement a large-scale
Condorcet-efficient system.
	I'd also like to point out that, while it will be difficult to elect
government representatives in the US using Condorcet's method, it might be
easier to adopt a pairwise tally in direct issue voting, if a new direct
issue voting system gets off the ground. The difference, as I see it, is
that there is less of a conservative backpull when one is designing a new
institution from scratch rather than attempting to reform an existing
institution. I have written more about this in my direct democracy
proposal, as well as elsewhere on EM. I intend to give the proposal a
major overhaul as soon as possible, but the current version can be found
here:
http://fc.antioch.edu/~james_green-armytage/voting_methods/proxy.htm

all my best,
James Green-Armytage
http://fc.antioch.edu/~james_green-armytage/voting.htm