[EM] 3ballot / rouse idea

Warren Smith wds at math.temple.edu
Mon Oct 2 17:45:20 PDT 2006


>Michael Rouse:
I originally thought that simply having two ballots reflecting my choice
(say, A>B>C) and one the anti-choice (C>B>A) would be sufficient to allow
a ThreeBallot version for Condorcet. After looking at valid criticisms of
this (thanks to Warren Smith), I realized there is a better way: One
authentic ballot, one random (or voter-picked) order, and one the reverse
of the random order.

For example, let's say my true preference is A>B>C. Some political
operative wants to buy the vote B>A>C. I would simply pick the following:

Ballot 1: A>B>C (true preference)
Ballot 2: B>A>C ("bought" vote)
Ballot 3: C>A>B (reverse of "bought" vote)

Now, I can print out ballot 2 as my "proof." I can pick any order I want
for ballot 2, and simply reverse the order for ballot 3. (Of course, I can
put the random and reverse random ballots at any position I want and my
"true" ballot in the remaining position.)

Of course, it would be rather tedious. If it were a computer/printer
combination, we could have a selection that would automatically randomize
one ballot and reverse this for another, and then print out the triplet.

Of course, I could be overlooking something, but it's an interesting
problem. :)

--that's pretty neat!
However, sadly, I still think it is busted.
The problem is the 3-pattern attack.
Coercer/buyer says (in a 33-candidate race like they just had in the Congo)
"I want you to vote A>B>C>...>Z as your `true' vote,
and X>Q>F>...>B and its reverse as your fake vote."
There are 33! votes possible (>> population of Congo)
so coercer can uniquify easily.  Then Coercer, if he does not see all three
demanded patterns posted on the bulletin board, exacts horrific
punishment on you, such as making you listen to a Bush press conference.

So you see the tremendous devastation this whole 3-pattern attack idea can wreak.
But in range and approval voting, all candidate scoring decision are independent
and can be on seperate "ballots" so there IS NO pattern, so not a problem.
With Condorcet, Plurality, etc, they depend and cannot be seperated
without making ballot-validity-check impossible.

Warren D Smith
http://rangevoting.org



More information about the Election-Methods mailing list