[EM] Proportionality in perspective

[LAYTON Craig]

-----Original Message-----
From: Blake Cretney [mailto:bcretney at postmark.net]
Sent: Monday, 12 February 2001 6:44
To: election-methods-list at eskimo.com
Subject: Re: [EM] Proportionality in perspective
n Sat, 10 Feb 2001 10:35:30 -0800
Bart Ingles <bartman at netgate.net> wrote:

>One point you mention is "locality of representation." 
>Leaving aside the issues of proportionality, and list vs.
>individual systems, I would argue that locality of
>representation is actually a bad thing, and it certainly
>doesn't make sense to compromise some other standard in
>pursuit of it.
>
>You may have some justification for locality.  I can't
>recall ever having read someone directly defend the
>principle.  People usually just assume that it is a good
>thing.  However, on the surface it isn't obvious why we
>should be representing people on the basis of geography.

In Australia, and I assume all Westminster-based systems (you can correct me
if I'm wrong), the local member performs an important purpose - "making
representations" on behalf of constituents in their dealings with government
bureaucracy.  If there weren't local members, I'm sure that some alternative
could be found, but it works really well.  Despite party solidarity, local
members will still pressure ministers from their own party for a response if
your grievance appears serious and founded. This is likely to be even more
effective if you have a number of local representatives, so that you can go
to one of the same party as yourself, and there is an extremely good chance
that your first or second choice is holding office.

But, as I pointed out, there could be an alternative to this process.  What
there is no alternative for, is putting candidate election into the hands of
the voters.  It is an unevoidable consequence of party list systems that the
parties basically decide who will hold office before the election.  In
reality, only a handful of seats are ever contested (you would know, for
instance, that the first 40 republicans on the ticket and the first 40
democrats were going to be elected).  A smaller electorate allows you to
have candidates listed straight down the ballot, and not grouped into party
lists.  You can add the additional feature of rotating the ballot paper, so
that no democrat appears above another democrat on every ballot paper, and a
voter voting for all the democrats straight down the ballot (in STV), would
be cancelled out by other democrats doing the same on rotated ballots.  I'm
assuming that the advantages of being able to distinguish between members of
the same party, and making no seats "safe" are clear, but I could detail
them in a subsequent message.

Despite the fact that you can design list systems to allow people to
distinguish between candidates of the same party, in practice voters do not
take this option.  If you have a free list system, which makes the best
attempt of the list systems to give voters a choice, the result will be
determined by a small handful of voters, which will either be from a strong
minority faction, able to instruct it's supporters in detail how to vote, or
by only the most educated voters, again disenfranchising poorer, working
class voters.