[EM] Dave Ketchum reply, part 3

MIKE OSSIPOFF nkklrp at hotmail.com
Tue Jul 2 18:14:30 PDT 2002

I'd said:

>Doesn't every county have an at-large board of supervisors? In
>that case, each county has provision for accepting & counting votes
>for several candidates per ballot.

Dave replied:



Yes, though I knew that boards of supervisors are usually elected
the same way, I for some reason forgot which way that is. Of course
they're usually or always elected in single-member districts, so as
to give local representation in the relatively large county

But lots of cities elect their councilmembers at-large, and so they're
set up to receive and count, on each ballot, votes for several
candidates, recording and incrementing vote totals for all the

True, many cities elect their councils in districts, but there are
so many at-large cities that it surely wouldn't be difficult for a
county govt or a district city to obtain what it takes to count
on each ballot votes for several candidates, recording all their
vote totals, as the at-large cities already do.

Dave continued:

Anyway, question is whether equipment built 60-70 years ago can be used in
a way its makers may not have planned on. You DO NOT KNOW unless someone
has actually tried that use and/or is familiar enough with actual machine
design to answer based on that.

I reply:

True, I don't know whether or not jurisdictions that have only been
doing district elections can use their equipment to count votes for
several candidates on each ballot in a particular race.

I'd said:

>But on this list, about half of "Condorcetists" are actually
>Margins advocates. If your Condorcet proposal is going to result
>in a margins vs wv fight, forget about it and propose Approval instead.

Dave replied:

THIS IS a significant topic. Best I can propose right now is to lock both
groups in a closet until they get their differences sorted out in private.

I reply:

I don't know how successful that would be. The Condorcetists and
the Margins advocates prefer those different methods because they
have completely different standards that they'd like a rank method
to meet. I'm afraid that it's going to have to be up to the public,
or maybe the initiative committees, to choose which standard means
more to them, except when the Condorcet proposal isn't opposed by
advocates of a different Condorcet version. I wouldn't oppose other
Condorcet version count rules, though I'd oppose Margins. Of course
there's always the possibility that a Margins proposal could sneak
through furtively, if it isn't mentioned on EM and no one e-mails me about 

Your suggestion is similar to what I suggested to CVD, that single-winner 
reform advocates have a collective discussion about what
would be a good single-winner reform proposal. No such luck. Those
who run CVD had already chosen their proposal and wouldn't reconsider
it, regardless of its unpopularity among those who study single-winner

Again, it's difficult to say whether Approval or Condorcet would be
more winnable. People initially seem more enthusiastic about rankings,
but then we have the issue of how to count them. And do we gain more
winnability by appealing to the progressive organizations that have
already embraced IRV, and therefore like rankings, or by appealing to
the voters, who might prefer the simplicity, minimal change, and
lower pricetag of Approval? I don't know. And, not knowing, I'd be
inclined to propose Condorcet(wv), or, more specifically,
Ranked-Pairs(wv). Maybe proposing it is the only way to find out its

When replying to an article or letter advocating IRV, it certainly makes
sense to offer, as a better alternative, a better way to count rankings.

One beauty of Approval is that you can tell the whole proposal and
count rule to someone in a few brief words at a bus-stop, etc. I
tried the RP(wv) definition, from scratch, on a friend, starting with
rank balloting, then pairwise-counting, and then the RP rule. I also
told them Approval. I asked which explanation they preferred, and they
preferred the brief Approval definition.

But sometimes of course you have time, and the others will listen,
for defining Condorcet. When proposing Condorcet in reply to IRV
letters or articles, I suggest one paragraph showing how IRV leaves
preferences uncounted, with disastrous results. Then a paragraph
pointing out that pairwise-count counts what the rank voter wants
counted--we want rank balloting so that we can vote, and have counted,
any preferences between any pairs of candidates. Then mention that
when the public's pairwise statements conflict, we obviously
can't honor all of them, and so the obvious solution is to honor the
stronger ones, and that leads right to the instructions for Ranked-Pairs. As 
I said, for public elections, I like RP's brief definition.

Of course, instead of guessing, or proposing Condorcet because we
don't know whether Condorcet or Approval is more winnable, it would
be better to conduct some sort of poll, between Approval and Condorcet.
Ask people which they'd be more inclined to sign and vote for an
initiative for. Or show each person Approval OR Condorcet, to find
out which gets better acceptance. If you ask people on a streetcorner,
I can tell you now that Approval will probably do better, since
people are unlikely to stand still for a rank-count definition.

But if you ask organization leaders and members, Condorcet would
have a much better chance. Then, with local progressive organizations
behind an initiative, the man-on-the-street might accept Condorcet
for that reason, or at least be more willing to listen to its
definition. I emphasize that RP's definition is briefer than the IRV
definitions that CVD uses.

Mike Ossipoff

Join the world’s largest e-mail service with MSN Hotmail. 

For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em

More information about the Election-Methods mailing list