[EM] (no subject)

James Gilmour jgilmour at globalnet.co.uk
Sun Nov 9 15:28:01 PST 2008


Dave Ketchum  > Sent: Sunday, November 09, 2008 10:59 PM
> I have not inspected the affidavits for completeness or 
> correctness.  I am only comparing the methods.
> 
> Assuming IRV's rules result in declaring A or B winner, it would not care 
> or look at what this voter may have said about C or D.
> 
> Condorcet looks at all that the voters say, and uses all of that in 
> deciding on a winner - as to C and D the possibilities are:
>       C>D
>       D>C
>       C=D = the voter indicates equal liking by giving them 
> the same rank or by ranking neither.

There is only one legitimate interpretation of the "A>B" ballot paper in a Condorcet count with regard to the "C" vs. "D" pair-wise
contest  -  the voter has given the Returning Officer no information.  No-one is entitled make any supposition  -  that voter has
expressed no preference at all as between "C" and "D".

However, all of this is totally irrelevant to what is in the affidavits and what my question was about.  In the affidavits it is
asserted that because IRV would treat differently the ballot papers marked "A>B>C>D" and "A>B", this is ONE of the reasons why IRV
counting should be declared "unconstitutional".  However, some of those who have taken this position, have in posts to this list,
indicated that they would accept Condorcet counting.  But Condorcet counting would also treat these two ballot papers differently.  

That leaves me genuinely puzzled as to how one such difference could be "unconstitutional" but the other not.  This is a very
important question because if IRV is held to be "unconstitutional" on THIS ground, then a whole raft of other voting systems,
including Condorcet counting, would also have to be considered "unconstitutional".

James Gilmour
No virus found in this outgoing message.
Checked by AVG - http://www.avg.com 
Version: 8.0.175 / Virus Database: 270.9.0/1777 - Release Date: 09/11/2008 09:53





More information about the Election-Methods mailing list