Truncation Resistance #2 criterion (was Re: First Choic

[Steve Eppley]

Hugh T wrote:
>Steve E wrote:
-snip-
>> Truncation will occur even with low deviousness (and maybe
>> especially with low deviousness), but massive reversal requires 
>> a highly devious group.
> 
>This does not make sense to me.  Truncation as offensive strategy
>requires more, not less participation because each ballot is less
>potent.  Order reversal can succeed with only a minority of the A
>voters participating. It does not seem more devious to withhold a
>true preference 
-snip-
>I am puzzled by the notion of "guarding" against "innocent"
>truncation. (Perhaps the system also should "guard" against
>complete abstention?)
-snip-
>If they were really indifferent between B and C (which is the only
>rational explanation for innocent truncation)
-snip-

It doesn't matter whether we all can agree that innocent truncation
is rational or not.  What matters is whether truncation of a
compromise candidate (who would win if voters completely express
their actual preferences) would be a significant occurrence.  
Mike O has asserted there is empirical (anecdotal) evidence that
non-strategic truncation will occur; perhaps we should attempt to
gather more conclusive evidence if the point is really in dispute.  

In addition to the anecdotal evidence already posted, there are
reasons to believe many voters will truncate their actual unequal
preferences without intending "offensive strategy":  
1) It takes extra labor to rank more candidates.  If many candidates
  are competing, voters will be inclined to "draw the line somewhere" 
  and may fail to rank as far as the compromise candidate.  The 3-
  candidate example may obscure this because it is so simple--so 
  consider an election where in addition to the 3 viable candidates, 
  the voter with the ABC preference order in addition happens to prefer 
  many nonviable candidates (D,E,F,G,H,I) more than B:  DEFAGHIBJKLC
2) A voter may (irrationally?) believe s/he should not rank some 
  candidate (in particular, the compromise candidate) if the candidate 
  fails some litmus test dear to the voter, even if the voter prefers 
  that candidate to others.  We've presumably seen this behavior 
  already on a massive scale--many nonvoters choose to abstain 
  rather than vote for a "lesser evil."

Perhaps Hugh doesn't believe that truncation of actual preferences
is bad for society, and doesn't mind if the voting method actually
creates an incentive to truncate rather than resist the tendency
of truncation skewing the majority preference.

I'm not sure where Hugh was headed by suggesting, probably
rhetorically, that to be consistent maybe we should try to find 
a system which guards against "complete abstention."  We can already 
say that Condorcet & Smith//Condorcet guard against abstention by
the minority, and that no system can guard against abstention by the 
majority.  Furthermore, we can speak of incentives: by eliminating, 
for practical purposes, the spoiler dilemma faced by potential 
candidates, Condorcet & S//C maximize the number of choices 
(viable and symbolic) available to the voters.  If the decisions 
of eligible voters whether or not to vote depend upon the "ballot 
presence" of "near" candidates, or on a system which minimizes 
headache-inducing strategy considerations, then the incentives to 
abstain are minimized. 

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