[EM] Duverger's Law

[Bart Ingles]

Forest Simmons wrote:
> 
> I read the interesting paper referenced below, but couldn't tell which of
> Duverger's observations is supposed to be "Duverger's Law."
> 
> Is it just that a first past the post lone mark single ballot system tends
> to keep a two party system entrenched?

That's my understanding.  The site where I fould the link made the point
that Duverger himself didn't coin the term "Duverger's Law", so that it
probably wouldn't do to search for an exact definition in Duverger's
writing.


> Here in the USA the two party system has degenerated into more of a one
> party system, with superficial cosmetic differences.  Is this also a
> natural consequence of the voting system?  Has anybody written on the
> dynamics that lead to one party systems?

I've seen the observation in a lot of places, but there are apparently
plenty of people out there who think the differences are anything but
cosmetic.  Otherwise, why would anyone care who won Florida?

Part of the problem may be that a two party system only permits a
one-dimensional issue space (i.e. liberal vs. conservative).  So long as
your own views can be translated to a position on this single axis, the
choice will seem material.

Issues which translate to "neutral" on the lone axis will tend to be
ignored, or at least have no direct effect on the outcome.  Although I
think candidates tend to use these to fine-tune their campaign
strategies.

The one-dimensional universe seems to be the most stable under the
current system.  A zero-dimensional, one-party system would invite
opposition, while a multi-dimensional system collapses down to one. 
Although zero dimensions would be stable under a strict dictatorship.  I
wonder if there is a voting system where exactly two or three dimensions
are stable?

I assume Approval or Condorcet would permit n-dimensions, but I wonder
about the value of n.  Maybe n would be infinite, but with some
dimensions "larger" than others.  The realities of campaign finance and
public attention might limit the value of n, or make it subject to
factors other than the voting system.  For example, liberal public
financing might increase n, while strict limits on campaign advertising
might force a fixed number of dimensions to have equal weight (I'm not
saying either of these are good or bad).

Bart


> It seems to be a corollary to Duverger's Law.  The two parties that
> survive Duverger's Law compete for the perceived median voter position on
> superficial matters while both strive to maximize satisfaction for their
> capitalist backers.  [The same big money donors support both parties in
> order to hedge their bets when the races are close.]
> 
> In other words, it's no accident that Gore and Bush were running neck and
> neck in the last presidential race.  The big money wants to reinforce
> Duverger's Law to make sure that whichever of the two parties wins, it's
> going to support corporate welfare.
> 
> Forest
> 
> On Mon, 24 Feb 2003, Bart Ingles wrote:
> 
> >
> > Just to further muddy the waters on the definition of "majority", note
> > Duverger's use of the term as apparently synonymous with "plurality":
> > http://www.janda.org/c24/Readings/Duverger/Duverger.htm
> >
> > (BTW the "two-ballot majority system" Duverger discusses is NOT the
> > Runoff method, since there is no fixed elimination rule.