[EM] Approval and PR (was Request for help: complex election)

James Gilmour jgilmour at globalnet.co.uk
Tue Mar 9 15:53:54 PST 2004


Jan Kok wrote:
> I think a "problem" (you may not agree that it is a problem) 
> with Approval voting for choosing multiple winners, is that 
> it is not independent of clones.  I.e. if there are 10 
> candidates who are all very similar, and all of them are 
> acceptable to a majority, and there are several other 
> candidates who are highly valued by several minority 
> factions, then the winners of the Approval election will 
> likely all be from the 10 "clones".
> 
> Some people support proportional representation (PR) methods 
> because they want to force the set of winners to somehow 
> "proportionally represent" the various factions of the electorate.
> 
> Do you want to enforce diversity (in the US it's sometimes 
> called "affirmative action"), or do you want to do the 
> opposite and create a situation where the winners may all 
> represent a single majority faction?
> 
> Although I have some sympathy for the idea of PR (because I 
> belong to a minor political party), I am also nervous about 
> having representatives of some _other_ minor parties get 
> elected.  Some of them could be worse (in my view) than the 
> mainstream candidates.
> 
> So, a tendency of an election method to choose most winners 
> from a majority faction may or may not be a problem - it's a 
> matter of values, not a clear-cut matter of right or wrong.

I don't think it is a matter of "values"  -  I think it is a matter of "purpose".

Q1: "Why are we holding this election?" A1: "To elect our town council".  Q2. "Whom is the town
council supposed to represent?"  A2: "The council should represent the electorate of the town, or at
least those who are prepared to vote".  Now we have a clearly defined purpose for our election.  Now
we have a criterion by which we can assess how well or badly different voting systems perform in
achieving that purpose.

You'll gather that I believe a body like a town council should represent all significant views
expressed by the voters in their responses to the candidates who offer themselves for election.  I
do not think any one group should have total control of the town council to the exclusion of all
significant minorities.  The purpose of the election is not to manufacture an apparent "consensus"
among the voters.  I realise this view may not be shared by those students of electoral methods who
come from a social choice background.  Where you are seeking consensus on a proposition (eg for the
allocation of resources) there is a parallel with the challenges of electing one single
office-bearer in a way that "best" represents the wishes of the voters.  But that social choice
theory, leading to manufactured consensus, has no place in the multi-winner elections, like the
election of a town council.  I reject dictatorship in all its forms, even when those collective
dictators have the support of a majority of the voters.  Significant minority views should be
represented, in due proportion.  Anything less is not democracy.
James




More information about the Election-Methods mailing list