[EM] Why We Shouldn't Count Votes with Machines

[James Gilmour]

> > Dave Ketchum wrote:
> > Mixed into this, Plurality is easily done with paper; better systems, 
> > such as Condorcet, are difficult with paper, but easily handled with 
> > electronics.

Kathy Dopp  > Sent: Saturday, October 04, 2008 1:24 AM 
> Well that is a very good reason to avoid implementing them - 
> because if they can't be easily done with paper ballots, then 
> they cannot be assured to be counted accurately.

This raises a very interesting point - how to balance the risk of failing to detect a low level of fraud against the known wasting
of very large numbers of votes by the plurality voting system.  (I say "low level of fraud", because any high level should be
readily detectable.)

Of course, we don't want any fraud, and we don't want any fraud to go undetected, and we don't want the outcome of any election
determined by fraud, no matter how low the level of that fraud may be.  But to use the ease of detecting fraud as the sole criterion
for selecting a voting system is almost certainly to lose sight of the much larger "losses of votes" that occur in every plurality
election.

In the UK, Canada and in most countries using plurality (except USA), the voting system discards the votes of around half of those
who vote - sometimes a little more than half, sometimes a little less.  In some plurality elections large numbers of the elected
members are elected with only a minority of the votes cast in the single-member districts.  The evidence on this is abundant and
worldwide.  The exception is the USA, where, for example, in elections to the House of Representatives, only one-third of the votes
are wasted in this way.  The reason is probably related to successful incumbent gerrymandering of the district boundaries and to the
effects of holding primary elections.  But even in the USA, around one-third of the votes are wasted by the plurality voting system.

So to look at the overall picture with a voting system like plurality, should we reject any move to a voting system that would give
effect to more of the votes actually cast because it might be more difficult to detect a low level of fraud in such a voting system?

James
No virus found in this outgoing message.
Checked by AVG - http://www.avg.com 
Version: 8.0.173 / Virus Database: 270.7.5/1706 - Release Date: 03/10/2008 18:17