Majority Support?

[Steve Eppley]

DEMOREP1 wrote:
-snip-
>There is a technology component to the various methods. Not all
>private groups or all governments at all times will have the
>technology and/or the patience to input high tech votes and do all
>the Condorcet pairings. Example- In Rwanda-Zaire with the mass
>killings, mass disease, mass starvation, etc. the governments/gangs
>presumably have a low priority in having a computer do a Condorcet
>election.

If Rwanda and Zaire don't have computers, then IRO will also be
difficult to tally.  

Plain Approval is better than the vote-splitting variation of IRO.
(Which is the one normally advocated by IRO advocates.  See the
discussion about the Instant-Runoff-1 variation, which doesn't split
votes among equally ranked first choices.)  Plain Approval would be
easier for the Rwandans to tally than IRO.  Yet plain Approval was
also omitted from your response to the sw poll.

I think this is a very weak rationale for ranking plain IRO above
plain Condorcet in our poll.  I'll be sure to quote the above in our
Commentary, since it implies that you really prefer Condorcet more
than IRO in the elections which are significant to U.S. reformers.

>>I have a few questions regarding the Majority Approval standard
>>(and the similar standard, Majority Disapproval):  
>>
>>Case 1: No candidates are majority-approved.
>>1.1 Why do you think a good candidate who would be
>>majority-approved in a re-election can be found?  What barriers
>>would keep such a good "white knight" from competing in the first
>>election but not prevent him/her from competing in the re-election?
>
>I never brought up anything about disqualifying anybody for
>re-election. Not sure where you got that idea.

You misunderstood the question.  I'm basically asking: why would
good candidates choose not to compete in the original election if 
they're able to win the re-election?

>>1.2 Are you concerned that the voters would "tactically" or
>>"capriciously" vote to majority-disapprove all the candidates even
>>though there's a decent compromise running?  
>>For instance:
>>   30:  A|BC                 <-- these disapprove the compromise B
>>   15: AB|C
>>   10:  B|
>>   15: CB|A
>>   30:  C|BA                 <-- these disapprove the compromise B
>>
>>Enough voters might opt to have no winner indefinitely rather than
>>compromise, and ensure no one will ever win.  Maybe some voters
>>will prefer anarchy over compromise. 
>
>All 3 do not have majority approval.

Right, that's the problem I'm pointing out, but you didn't answer
the question.  The question I asked was: ARE YOU CONCERNED that this
can be the outcome even when the compromise candidate is really okay?
Remember, voted rankings and disapprovals aren't always *sincere* 
preferences.

>>Case 2: At least one candidate is majority-approved (or would 
>>be, except that the "plain" ballots don't allow expression of 
>>approval/disapproval).
>>
>>2.1 Are you concerned that plain Condorcet might elect a
>>disapproved candidate when there's at least one approved candidate
>>running (e.g., Hitler vs Stalin vs George Washington)?  
>
>One cannot tell whether or not a candidate is majority approved
>until there is a vote. George Washington might be the B in the Hippo
>series example of 46 ABC, 20 B, 34 CBA with a 100 approval.  See my
>earlier Hippo response.

I don't know why you're missing the point of all my questions.  I'll 
try rephrasing:

Can you construct a realistic example where the winner by plain
Condorcet could become majority-disapproved if some | dividing lines
are inserted into the rankings, without making all the candidates
become majority-disapproved?

Here's an attempt to meet this "example spec":
   45: Washington Hitler Stalin
   20: Hitler
   35: Stalin Hitler Washington
    ==> Hitler wins by plain Condorcet.
   Could Washington have been elected if voters had also been able 
   to express disapproval of Hitler?  Insert some disapproval 
   dividing lines:
   35: Washington | Hitler Stalin
   10: Washington Hitler Stalin
   20: Hitler
   15: Stalin Hitler Washington
   20: Stalin | Hitler Washington
    ==> 55 voters now disapprove Hitler, but Washington is only
        disapproved by 20.
So the question is, is this scenario realistic enough to be of 
concern?  If so, it meets the spec.  If not, can you construct 
a realistic example so we have something to really be worried 
about?

>>If so, would you provide an example which illustrates this scenario?
>>If not, would you agree that Condorcet's counting of pairwise
>>preferences instead of first choices, and its selection of smallest
>>"largest votes against in a pair-loss", already does a good job of
>>selecting the majority-approved candidate, even when the voters
>>can't express approval? 
>
>A good job is not guaranteed unless there is a majority approval
>vote.

If there's no realistic scenario such as that referred to in my
rephrased question, then you're worrying unnecessarily, and since
insisting on the option to disqualify all the candidates will
definitely make the reform proposal much more controversial, it 
may be very unwise to do so.

>Simple disapproval indicates a rating below zero. 
-snip-
>Since percentage ratings cannot be used due to the tendency to rate
>candidates as +100 or -100, an approval/disapproval vote will only
>remove the below zero candidates.

Not true.  There's a lesser of evils dilemma in approval voting
which can distort approvals: should I approve a compromise candidate
and risk defeating my true favorite, or should I disapprove the
compromise candidate and risk electing the greater evil?  This can
result, for example, in the disapproval of a candidate who would
have been approved had s/he been the only candidate running.  
And it can result in the disapproval of all candidates even though
there's one who would have been approved if all voters had voted 
sincerely.

---Steve     (Steve Eppley    seppley at alumni.caltech.edu)