[EM] [ESF #1080] Re: Meta-criteria 3 of 9: Value: expressiveness

[Jameson Quinn]

2010/5/10 Abd ul-Rahman Lomax <abd at lomaxdesign.com>

> At 05:00 AM 5/10/2010, Raph Frank wrote:
>
>> On Thu, May 6, 2010 at 10:19 PM, Jameson Quinn <jameson.quinn at gmail.com>
>> wrote:
>> > Only when the vote is an
>> > election, not a poll, does dishonesty come into the picture, as utility
>> > conflicts with expressivity.
>>
>> However, as you say, choosing to vote approval style under range is a
>> statement too.
>>
>
> What, precisely, is "expressivity" if it is not the ability to express
> utility? As we have been emphasizing for a long time, voters will naturally
> include probabilities in the energy they put into a choice. That "energy"
> has various aspects: the effort involved in learning about the choices
> involved in an election, showing up to vote, and how they use their voting
> power. They may use their voting power to "make a statement," entirely apart
> from what the expected results are. Or they may restrict their expressed
> choices to those they think might actually affect the outcome.
>

Expressivity is precisely the ability to express one's true preferences in a
way that will have a meaningful and congruent impact on the aggregate
results. This is not the same as only caring about who wins, and voting
strategically for best effect there. An expressive voter would care about
the relative strengths of a number of candidates.

Any aspect of a voting system which can impact the outcome is immediately
subject to some anti-expressive strategy. Therefore, if you want to maximize
expressive power, it may be worth having some "polling" aspects to the
ballot, which have no impact on the result, and are exclusively useful for
expressiveness.

We do not put effort, most of us, into useless gestures, mere "expression."
>

Millions of people vote for third parties in any US election. Tens of
millions more vote in the "safest" of races, even though they know the
result perfectly well before voting. These actions can only be justified as
expressive ones.



> Now, it seems that some imagine that the ideal voting system will extract
> from voters some kind of "sincere expression" of utility, in order for
> utility to be useful;
>

What I said was much more precise than that. Utility is only one of the
possible values. Even if utility is the main goal, in order to analyze the
utility of a system you need some model for underlying utilities and for
strategic behavior. If you have no model for the circumstances under which
voters are strategic, and only count them as fundamentally inclined for or
against strategy, then cardinal systems (Range), which count utility
explicitly, give the highest-utility results. If your model for strategy is
based on cabal strategies, which are only dependent on the ordinal
preferences, then perhaps ordinal systems do better.

When you set out what "some imagine", please try to unpack the ideas a
little bit more precisely. Otherwise, you end up with uselessly-vague straw
men.

>
>  The ideal system would be something like declared strategy voting,
>> where you just tell the system your utilities and let it worry about
>> the back-end.
>>
>
Unfortunately, optimal strategy in the situation of a cyclical tie is a
mixed strategy. That means giving probabilistic results in this situation.
Most people don't like that, and most DSV proposals aren't actually
probablistic. For instance:


>
> This is actually how Bucklin works.


...my point.

Actually, optimal strategy for the Range/Bucklin election you propose would
be something akin to optimal Range strategy. The strategic incentives are
significantly less than they are with Range, so I do like your proposal a
lot, but you can't call it DSV. With Range, the incentive is to be as
strategic (approval-style) as possible; with this system, the incentive is
to be a little bit more strategic (approval-style) than average. Consider
the case where you know all other voters except one are voting
approval-style. You want your vote to be at least one rank more
approval-style than that other voter; you want to compromise with others
(high-side) one step before them, or compromise with them (low-side) one
step after they compromise with you. So the safest vote is pure
approval-style.


> But it would be easy to vote, easy to instruct voters how to effectively
> manage their ratings on the range ballot.


Nope. Strategy is still hard, and it still has some payoff. But you could
maybe convince people that the payoff was small enough that it wasn't worth
it. So I do like the system. But it isn't DSV.

>
>  > This leads to a certain paradox: systems which seek to increase
>> > expressiveness by increasing voter freedom - for instance, Range as
>> compared
>> > to a Condorcet system - could increase strategic opportunities, and thus
>> in
>> > the end reduce expressiveness - for instance, if Range were to end up as
>> > pure Approval in practice.
>>
>
>
>  That isn't a paradox.  Range is more expressive than approval even if
>> 99% of the voters vote approval style.
>>
>
Sure. But is it more expressive than ordinal systems (Condorcet), or than
(standard limited-rank, equality-allowed) Bucklin, in that case? I think
not. Yet the ballot is more expressive than those alternatives. So it IS a
paradox: the more expressive ballot could, in practical reality, give the
less-expressive system.


>
>  Expressiveness isn't reduced, it just isn't increased as much as it could
>> be.
>>
>
> That's right. Reduction in expressiveness from bullet voting is actually
> not a reduction, it's an *expression.* It expresses strong preference.


Not if that's not what the voter wanted to express. (The fact that it's the
outcome the voter wanted does NOT imply that it's what they wanted to
express.)


JQ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20100511/2cfb0e94/attachment-0004.htm>