Reply from Lorrie Cranor on Neutrality

[New Democracy]

Greetings to Election Methods List,

     On January I posted on this list a letter that I sent to Lorrie Cranor
titled: Approval Voting and Borda Count are not neutral. I have received an
answer from her - which I am posting below.

Donald
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
From: Lorrie Cranor <lorrie at research.att.com>
Subject: Re: Approval Voting and Borda Count are not neutral
To: donald at mich.com (New Democracy)
Date: Wed, 15 Jan 1997 10:09:38 -0500 (EST)
Organization: AT&T Labs-Research
MIME-Version: 1.0

Thank you for your comments.  I agree with most of your analysis
regarding neutrality; however, I still maintain that approval and
Borda count voting are neutral.  My use of the word "favor" in the
definition of neutrality refers to favoring an alternative by virtue
of something external to the actual distribution of preference.  Thus
the amendment procedure favors the alternative representing the status
quo, regardless of what the actual distribution of preference is.
I believe my definition and interpretation of neutrality is consistent
with most of the voting literature.

The point you raise is interesting, however, and suggests that another
criterion of voting systems be considered.  I'm not sure exactly how
to define this, but it should have something to do with what the
results of an election represent.  We need to make a distinction between
the ability of a voting system to tell us who the winner is, and the
ability of a voting system to provide information about voter preferences.
But the ability of a voting system to provide preference information
goes beyond the totals it produces.  For example, in my work on
Declared Strategy Voting, I discovered that DSV election tallies can
often be misleading when viewed alone.  However, when viewed with
various intermediary results produced by the system, they can provide
very rich information about preferences that is not available from
systems that produce more representative tallies.

Anyway... about Coomb's method.  I would have to look it up again to
give you more detailed information.  I recommend you take a look at
Comparing Voting Systems by Hannu Nurmi.  Published in 1987 by
D. Reidel Publishing.

Regards,

Lorrie Cranor

-----------------------------------------------------------------------
Lorrie Faith Cranor                             lorrie at research.att.com
Public Policy Research, AT&T Labs-Research                 908-582-7914
600 Mountain Ave., Room 2C-430A                        FAX 908-582-4113
Murray Hill, NJ 07974              http://www.research.att.com/~lorrie/


- - - - - - - - - - - - - - - - - - - - end - - - - - - - - - - - - - - - -

Donald Eric Davison of New Democracy at http://www.mich.com/~donald

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -