[EM] SODA rationale, part 2 of 4: politicians and LNH (was: Record activity on the EM list?)

[Jameson Quinn]

>
>
>
*Continuing to consider SODA's advantages with groups often skeptical of
reform....*

Second, there's *politicians*. By definition, these are people who have
prospered under the current system; whose livelihood depends on knowing
their way around it; and whom you need at least some of on your side if
you're ever going to pass any reforms. If they were purely self-interested,
it would be pretty much impossible; because why would they want to change
the system that has made them winners? But, while you could certainly get me
to say all kinds of bad things about politicians, it's still important to
recognize that they're not just motivated by self-interest. All but the most
corrupt care at least to some degree about the public interest; and even the
most corrupt have to try to fake it.

So, imagine you're selling voting reform to some basically-honest,
public-interested politician. You have two tasks: convince them that it will
help the public, and help them understand how they could continue to prosper
under this new system. For helping the public, you talk about results; I'll
make some of those arguments, about how SODA gives good results, below.
Here, I want to talk about how SODA, unlike other systems, is not too
radical a change from a politician's point of view.

Under plurality, a politician's job is to be seen as the strongest candidate
in their own party, then to be seen as the lesser evil by 51% of the voters
at large. That is, they're affiliated with a party, and they're probably not
interested in changing that label. Voting reform is a threat if it makes
parties irrelevant, or worse, if it makes that (R) or (D) next to their name
into a liability.

This is where I listen to FairVote. Say what you will about their
sometimes-sneaky tactics (I find their recent false-flag blog at
rangevoting.com to be despicable, not to mention flat-out wrong on many
points), people like Rob Richie have by far the most experience promoting
voting reform with U.S. politicians. And I don't think their hangup about
Lesser-No-Harm (hereafter just LNH, becase they never talk about
Lesser-No-Help) is purely irrational. I think LNH matters to politicians.

Under non-LNH systems like Condorcet, if you're the second choice of three
or more non-majority groups, you win. To a politician steeped in plurality,
that seems like cheating. It means that any old centrist can come along,
without paying their dues in a party, and as long as they're positioned in
between the two major parties, they win. To put a name on this story, I call
it the Perot effect. I'm not claiming that Perot actually would have been a
Condorcet winner - at the time of the elections, at least, he probably
wouldn't have. But in the story, there's someone who is ideologically around
the median voter, and in a partisan electorate and a non-LNH system, they're
unbeatable.

LNH prevents that. If you've paid your dues, and you have status in a major
party, an LNH system will not look past that unless it has to. That's
important to a politician, who has already paid those dues. And it also
arguably has an element of public interest. Perot's quitting and
re-launching his campaign, in my opinion, showed a dangerous lack of
constancy, a quality that I think is important for a president. And the
party grind, the slow climb to status in a party, is one way to demonstrate
constancy and other qualities. (Personally, I don't think that all the
qualites it demonstrates are desirable. But they do include good qualities
like experience, constancy, and diplomacy). LNH guarantees that unvetted
candidates can't win in a DH3 pathology.

*And SODA satisfies LNH *for voters [1]*.* On the approval side, that's just
on a technicality. Since, simply going from the ballot, no approval is
"later" than an other, the fact that they can "harm" each other doesn't
break LNH. But on the delegated side, SODA is really squaring the circle,
providing a method that is both LNH and Condorcet compliant. It can
accomplish that impossibility partly because of the constraints on delegated
votes (they must follow one of the pre-announced rankings, each of which is
headed by a different candidate), but mostly because in the post-election
assignment phase, all totals and all rankings are known. A candidate might
wish they could conceal their rankings from earlier players in the
assignment order, but when their turn to assign comes, later-ranked
candidates cannot harm earlier-ranked ones. So a delegated vote, with the
candidate's pre-declared "later" rankings implicitly included, cannot be
worse than an undelegated bullet vote, unless you disagree with those later
rankings.

[1] It doesn't satisfy LNH for the candidate's initial declaration of
preference rankings. But a candidate who refuses to declare their second
preferences, simply encourages their voters to do so for them, which is not
in the candidate's interest. Only a frontrunner candidate --- that is,
someone who's been vetted, not a dark horse --- might get away with that,
because voters would not see the need to include backup votes.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20110803/6bd9843b/attachment-0003.htm>