[Membership] [Comment-Mac] Re: Secret ballots

[Kent Crispin <kent@songbird.com>]

On Mon, May 17, 1999 at 07:17:27AM -0400, Jonathan Zittrain wrote:
> Kent,
> 
> The problem with roll call as I see it is that it permits vote
> buying--"contracts" can be formed with verification that a vote has been
> delivered.

That is a theoretical possibility, yes.  

However, the protection secret ballot gives you is actually
fairly weak: votes can be bought or coerced on a probabilistic basis,
as well -- you don't need a verified 100% return on your investment 
to significantly influence the vote.  That is, you can buy votes or 
coerce votes in a secret ballot with only a marginal increase in 
effort and expese.

> If a secret ballot is used it's much more difficult to buy (or
> coerce, for that matter) votes.  ...JZ

I would say "it is *marginally* more difficult" -- eg "If I don't win
the election, your family will suffer" is effective, even without
100% verification of individual votes.

Note also that secret ballot is just as much a method for
*concealment* of vote buying/coercion -- when someone publically
supports a particular position, but secretly votes against it, that
is a strong indication of some form of double dealing.

Note that in the area where vote buying or coercion are the most
likely to be effective -- namely, at the Board level -- we *insist*
on open roll-call voting.

On the other hand, there is no doubt that with secret ballot bribing
vote collectors/counters is a much easier task than trying to
buy/coerce hundreds of individual votes, and likely to be far more
effective.  This is especially true in our particular case --
membership will be very widely distributed, most of the members will
never meet in person, and how will you assure a suspicious someone in
a remote region that the election committee hasn't been bought?

With public posting of the results, the remote member can 
still be suspicious of the posted results, but he knows with a high 
degree of certainty that his vote has been counted correctly, 
that everybody else is seeing it, and that if the votes were 
miscounted there would be widespread public complaint.  

In summary, a guaranteed honest secret ballot would be preferable to
a roll-call vote.  But the precise problem at hand is guaranteeing
honesty, and it is far, far easier to do that with a roll call.  In
the particular environment in question -- highly distributed
multicultural international elections -- construction of a secret
voting procedure that everyone will trust will be a major chore, and
probably fairly expensive.  Constructing a roll-call based voting 
procedure that everybody will trust is very much simpler.

A few other responses: 

Re Dr Nii Quaynor's concern about powerful people coercing votes: 1)
In principle I think it will be hard to avoid that whether secret
ballots are used or not.  2) In practice, however, I'm not sure that
this is a realistic concern.  A powerful individual in Nii's country
might indeed coerce Nii's vote.  But he isn't going to be able to
coerce my vote.  In general, given the very widely distributed nature
of the membership, it would be *very* difficult to coerce more than a
small fraction of the voters, and consequently, the cost-benefit
analysis would be that it isn't worth the trouble to try. 

If this is seen as a serious problem, however, a way to partially
alleviate it would be to allow members to set up a private identifier
that they could use to vote.  In extreme cases a properly identified
individual who requested anonymity could be granted some form of
alias or member identifier that masked their identity, even in
email...
 
Re Joop's point about electoral fraud coming from the top: That is
not an argument for secret ballots.  Indeed.  That is precisely the
problem *with* secret ballots -- they nourish all kinds of fraud at
the top.  In particular, the danger of a dishonest election process 
is very much lessened by roll call voting.

Re Diane's point: I would be interested to know why she thinks the
individual members privacy is so important, while the Board members
privacy is not.  All the potential abuses that secret ballots
presumably guard against are present in spades at the director level
-- vote buying and coercion are likely to be most effective when
applied to directors.  Why don't the demands for openness and
transparency come down to the individual member level? What,
precisely, is the value of being able to conceal your vote, and why
is it valuable at the individual level, but not at the director
level?

-- 
Kent Crispin                               "Do good, and you'll be
kent@songbird.com                           lonesome." -- Mark Twain