Saari reply

[Bart Ingles]

Dave Ketchum wrote:
> 
>       While Condorcet's cyclic ambiguities can be a debate topic,
> hopefully these only occur in near-tie situations, and those who would
> debate hopefully can be locked in a closet until they can agree on a
> public position.

No, sorry, there does not need to be a near-tie in order to have cycles.

Furthermore, you don't need to anticipate a cycle in order to have
incentive to truncate or rank insincerely.


> AND, I see explaining Condorcet to voters as reasonably simple:
>       Think of the candidates in order, starting with your top preference
> and continuing until you lose interest - and indicate this ordered list on
> the ballot.
>       Ignore any troublemaker who talks of spoilers.  Listing your desired
> minor candidate first will have ZERO effect on what your voting as to the
> major candidates does (of course, if enough voters show preference for
> that minor candidate, then that candidate would really be major and
> possibly win).
>       Ignore troublemakers who talk of desiring to show the same
> preference for two or more candidates.  You cannot do this, but are little
> deprived - ranking two candidates next to each other gives them the same
> preference relative to all other candidates - which you list first will
> not matter unless they are tied for winning.

This sounds like CVD's strategy for promoting IRV -- i.e. lie to the
voters, telling them that "it is always advantageous to rank candidates
sincerely", "the spoiler problem is eliminated", etc.

Granted, the likelihood for needing strategy is less for Condorcet
(except in certain situations), but it is not eliminated.  The only
voting system I know of that eliminates strategy is "random ballot" or
some variant thereof.

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