Eye Ball Mathematics
dfb at bbs.cruzio.com
Mon Nov 18 06:57:39 PST 1996
donald at mich.com writes:
> Dear Election Methods List,
> Thank you DEMOREP1 and Mike for answering the queries of my post
> called Eye Ball Mathematics. I have studied both of your replys.
> I now understand how this will work in a supposed election:
> Suppose I am able to talk my city into using the Condorcet method for the
> election of mayor. The election is held and we have a Condorcet winner. If
> the officials of the city ask me to prove that this winner is the correct
> winner I do the following:
> To verfity that this winner meets the Majority Rule Standard I merely
> use head to head pairing on the original election numbers to produce a
> candidate that is the indicated majority wish of the voters. I then compare
> this candidate to the Condorcet winner candidate of the election. If the
> two candidates are the same this proves that Cordorcet elects the candidate
> that is the majority wish of the people.
I don't know what you mean by that. If you could write things that
mean something, that would help you to get a better idea of what's
going on, in addition to giving us a better idea of what you mean.
As it stands, I don't belive that you know what you mean, and
maybe you don't mean anything.
> Recap - -
> We have a candidate that is the Condorcet winner in the race for mayor.
Do you mean a candidate who beats each one of the others, or merely
a winner by Condorcet's method. The two aren't the same thing
when there's no one who beats each of the others.
Already, your re-cap is getting off on the wrong foot, Don.
> We have another candidate that is the "implied majority" wish of the voters.
Again, I have no idea what you're talking about. I suspect that you
> We determined this second candidate using head to head pairing as suggest
> by DEMOREP1
> The second candidate becomes the "Majority Rule Standard".
A candidate is a majority rule standard??
> If we wish to prove the validity of the Condorcet method we need only to
> compare these two candidates - if they are the same candidate then this is
> proof that Condorcet elects the majority wish of the voters.
I'm afraid that you lost me way back there. Again, I don't recognize
any of this as Condorcet's method, or anything remotely resembling
it, or as anything remotely resembling our discussion of Condorcet,
or our explanation of the standards that it meets.
To the list owner: This person is wasting our time. He
doesn't show this list the respect of first taking the time
to find out whether or not he knows what he means before
he posts his meaningless strings of words here. Is there
a point at which someone like that is finally removed from
> But - -
> Is not head to head pairing the same as Condorcet? I am confused.
Yes, you are. Will you please figure out what you're trying to say
before you post it??
> I have a problem that I hope you are able to answer.
> Let me use the same supposed mayor's race: My city will use Condorcet to
> elect the mayor. But - the head of the election commission wants me to
> present to her, before the election, the following written materials: She
> wants the method and/or procedures and/or mathematics on how to vertify
> that the Condorcet mayor elect will in fact be the majority wish of the
> people. She wants this information so that she will be able to vertify for
> herself sometime after the election. She is a control person - she wants to
> know everything - she thinks she is in charge.
> She has no problem with the Majority Rule Standard. She has a problem with
> the determination of this standard. She is asking - How is it determined
> which candidate becomes the Majority Rule Standard for each election?
No candidate becomes the majority rule standard for the election.
Majority rule itself is a standard. We've heard the IRO version
of what majority rule means, and hopefully you've read one of
my postings in which I talked about a basic democratic principle
that's about majority rule. Do you really want me to repeat it
> What do I tell her?
Ok, then I'll repeat it again (My apologies to everyone else
for having to repeat this to Don again):
If a full majority of all the voters indicate that they'd rather
have A than B, then, if we choose A or B, it should be A.
If your election manager wants to know in what sense Condorcet
carries out majority rule, then you can tell her that it's
the only method that won't avoidably violate that basic democratic
principle. You can show her examples, like the 40,25,35 example,
or my recent 35,32,33 example, where IRO blatantly fails that
You can point out to her where, in that particular election,
majority rule is observed.
Of course if 1 candidate beats each one of the others, then
you'll have no problem.
I don't know exactly what she expects you to show her, to tell
the truth, when no 1 candidate beats each one of the others,
Maybe it's an election where there's only 1 candidate who
doesn't have a majority against hir. Then your job is easy.
Otherwise, you can tell her that the winner is someone
over whom the fewest voters have said they'd rather have
someone else who beat hir. That, in itself, is reason to
pick that winner. But if you're asking me to tell you a
majority rule standard that will always point to a single
candidate in every election--forget it. There's no such
thing. Sometimes there is, even when there's no Beats-All
candidate, but not necessarily. Once you asked why the
standard itself can't be used as the method. I answered that
it's because the standard doesn't always point to a single
candidate. But you're asking for a standard to do that,
and I'm saying it won't. So prepare to have that woman
throw you out the door. You said she wants you to show
her that the winner is the "majority wish of the people."
I'm saying that there isn't always 1 particular candidate
who can be so described, though majorities do often express
wishes, and when they do, the wishes should be honored--
as Condorcet will do, and as IRO so often will not do.
> I await your reply,
You got it.
> Your servant,
More information about the Election-Methods