2023-20 Presidents Gorbachev and Macron

In 1990, just after Moscow News published my article on consensus, Mikhail Gorbachev used this word, консенсус, for the first time.  And now another president is following my lead: Emmanuel Macron has just started to use the word préférendum. 

I hope it will be an MBC.

https://www.lemonde.fr/politique/article/2023/08/29/qu-est-ce-que-le-preferendum-nouvel-outil-democratique-envisage-par-le-gouvernement_6186971_823448.html

The following is now being translated into French...

A REFERENDUM…. OR A PREFERENDUM

A referendum is always binary,         

a majority vote,                                   either, “Option X, yes or no?” or a pairing “Option X or option Y?”

and the winner is usually the one with the most votes, 50% + 1, but sometimes it’s weighted.

A multi-option and therefore non-binary referendum may be analysed in a number of ways:  

a plurality vote                                  the winner is the option with the most votes… which might not be a majority but only the largest minority…

the two-round system, TRS             so the ballot could be followed by either a second-round majority vote between the two leading contenders, so that the winner does have majority support…

or the alternative vote, AV               a knock-out system: the least popular option is eliminated and its votes transferred until one option gets a majority; (the trouble is, a TRS majority winner may not be the same as an AV majority winner);

approval voting                                 is non-preferential.  Voters ‘approve’ of one or more options, and the winner is the option with the most ‘approvals’ – counting all the 1^st and 2^nd ‘preferences’, or maybe the 3^rd ones as well, or maybe also the 4^th ones too… which makes life confusing;

the Borda Count, BC                        is preferential.  Voters list (one, some or ideally) all the options in their order of preference, and at best, the winner is the option with the highest average preference.

The Condorcet rule                          is also preferential.  In a three-option ballot on options X, Y and Z, it considers all the pairings: X and Y, X and Z, and Y and Z, and the option which wins two pairings is the Condorcet winner.

So, by way of an example, consider 15 voters expressing their preferences as shown:

Preferences	NUMBER OF VOTERS
	5	4	3	2	1
1st	V	Z	W	X	Y
2nd	X	Y	X	Y	W
3rd	Y	X	Y	W	X
4th	W	W	Z	V	Z
5th	Z	V	V	Z	V

The outcomes are as shown in this next table:

Methodology		Social Choice	Social Ranking
Plurality voting		V	V-5	Z-4	W-3	X-2	Y-1
TRS		Z	Z-8	V-7
AV		W	W-10	V-5
Approval voting	1st & 2nd	X	X-10	Y-7	V-5	V/Z-4
	1st – 3rd	X/Y	X/Y-15		W-6	V-5	Z-4
BC		X	X-57	Y-53	W-43	V-37	Z-35
Condorcet		X	X-4	Y-3	W-2	Z-1	V-0

In effect, the result could be anything!  So just because a ballot is multi-optional does not necessarily mean that the outcome is accurate.  Indeed, as in this example and many other instances, the result depends not just on the voters’ preferences, but in some instances, on the voting system.  As Josef Stalin once said, “It’s not the people who vote that count, it’s the people who count the votes.”

Just looking at the first table, it is pretty obvious that options V and Z, are seen to be very divisive.

In contrast, some of the other options, like X and Y are both fairly acceptable, for neither of them gets any 3^rd or 4^th preferences. 

Of the voting procedures listed, only the BC and Condorcet take all the preferences cast by all voters into account, always.  They are the most accurate methodologies.  Little wonder then that they produce almost identical outcomes, with both X and Y getting very good scores.  {In a similar way, in a sports league, the champion, the team which wins the most matches (pairings) – the Condorcet winner –is usually the team which also scores the most goals (points) and the best goal difference – the BC social choice.} 

The BC is the more nuanced, and it’s non-majoritarian.  It is also robust, completely colour-blind, and extremely accurate.

Peter Emerson

Director, the de Borda Institute

34-6 Ballysillan Road

Belfast BT14 7QQ

+44-7837717979

www.deborda.org

pemerson@deborda.org

A Short Bio: 

Peter Emerson is the child of an English Catholic mum from Cheshire and an Irish Protestant dad from Cork.  He moved to Belfast in 1975 (from Africa) and was often asked binary questions: “Are you Protestant or Catholic?”  ‘Neither.’  “Are you British or Irish?”  ‘Both.’

In 1985, the two governments signed the Anglo-Irish Agreement.  The extreme Protestants met in Belfast and shouted, “Ulster says ‘NO’!”  So one week later six of us stood at the same venue, in silence, with a banner: “We have got to say ‘yes’ to something.”  Next we held a cross-community public meeting; over 200 came; anyone could make a proposal (if it complied with the UN Charter); then, on a ballot of ten options, they cast one or more preferences, saying ‘yes’ to something or somethings… but no-one voted ‘no’.  

The Cold War was another binary conflict, so in 1988 he went to Moscow as a translator and two years later wrote on consensus politics in Moscow News and Novy Mir (New World).  Back in Belfast in 1991, he organised another consensus conference using electronic voting.  The main guest was from Sarajevo, and six months before their war, he tried to say, please, not a binary referendum, not in Bosnia!

Having realised that this preferential points voting is a Borda Count, he now runs the l’Institut de Borda, promoting consensus voting at home and abroad, most recently in China.  His latest volume is The Punters’ Guide to Democracy, 2022, (Springer, Heidelberg).