Dear Alan,
"...the theory of voting... appears to be wholly unknown to anyone concerned with its practical applications... to the politicians... to others actually involved in decision-making... [and] largely unknown...to experts in political institutions." (The late Professor Sir Michael Dummett, one of the first patrons of the de Borda Institute.) Your Report comments on numerous details, but does not critique the binary nature per se of referendums past or proposed, be it with reference to Ireland or to Brexit. In a nutshell, your Report ignores the science. Why - or so it seems - is it mesmerised by what Michael* called "the mystique of the majority"? You're not the only ones of course. Many in academia and the media are equally uncritical. BBC Radio 4 did a documentary on referendums in 2011; I wrote to them beforehand about multi-option referendums, but in 60 minutes on air, not one word. Even the Electoral Commission did and does not consider multi-option voting - not for the 2011 referendum which was definitely multi-optional, not for Scotland in 2014 when there were three distinct options 'on the table', and not for Brexit (see below). The Greens too include those who feel that, while folks should be able to cast their preferences in elections, the very opposite should apply in decision-making. Furthermore, it's the same abroad: the Badinter Commission on Yugoslavia, for example, five Supreme Court judges, prompted the EU (the then EC) to "insist" that Bosnia had a referendum... it was (not the but) a cause of war. As in the GFA, the authors of Dayton also tied themselves into knots of majoritarianism. What's more, it has often been like this: the Parliamentary Council meeting in Bonn in 1949 introduced a new electoral system, but retained the much larger fault-line of Weimar, the binary vote of decision-making. Likewise, in 1776 in the USA, Jefferson and others were seriously looking at elections, but not at decision-making. This obsession with majority voting is indeed ubiquitous; it's even in Article 97 of the Constitution of the DPRK.
But let's return home. I think the biggest weakness of Brexit was its binary nature. The question was multi-optional. We had a binary vote on just one of the options. If we had had other binary votes on the other options - (which is how Slovenia chose their electoral system in 1996) - then the winner would have been the option with either a majority or, failing that, the largest minority; (Slovenia's winner got 45%, well ahead of the others on 26 and 19%). At maybe 48% in the Brexit poll, 'remain' would probably have been the option with the largest minority. But your Report critiques only other aspects of Brexit. As I say, it does not question binary voting per se, despite the latter being primitive and Orwellian.
Instead, your Report's "starting point is the Belfast/Good Friday Agreement 1998," (Chap 1 - 3), "[which] states: It is for the people of the island of Ireland alone, by agreement between the two parts respectively and without external impediment, to exercise their right of self-determination on the basis of consent, freely and concurrently given, North and South, to bring about a united Ireland, if that is their wish, accepting that this right must be achieved and exercised with and subject to the agreement and consent of a majority of the people of Northern Ireland."
And you conclude by saying (Chap 15 - 53) "...the GFA... stipulates that majorities of 50% + 1 would be required. But the ethos of consensual politics should be upheld as far as possible."
The first paragraph is confusing and confused; it is therefore, I suggest, an unreliable "starting point". Does the phrase "the people of the island" mean all of them, or maybe (in two separate constituencies) just 50% + 1? If the latter, then, in theory, the constitutional status could change by the decision of just two individuals, as what was 50% - 1 becomes 50% + 1. Or, though I hate to say this, such an outcome could be reversed by two pre-poll murders. {We have only to look at Bosnia to see how the prospect of a referendum had dreadful, horrible consequences, prior to the 1992 poll (let alone afterwards)}. Secondly, what is meant by a consent "freely" given? In the 2011 referendum on the electoral system, I was not "free" to vote for a system of PR, because David Cameron had restricted the choice to a dichotomy of AV or FPTP, which did not include PR; for me, the question was like that of the waiter who asks a vegetarian, "Beef or lamb?" Likewise, if a future referendum in Ireland is to be a dichotomy, it will inevitably mean that some people will not be "free" to vote as they would prefer to. Personally, I would like to see a W-I-S-E (Wales-Ireland-Scotland-England) confederation... but the Peace Agreement does not cater for such peaceful opinions. Thirdly, well, when two people consent to live together, the law requires the agreement of both; in contrast, in politics, it's different, and the GFA says that two communities may (can, must) live together, if the bigger one says so. Overall, then, the paragraph contradicts itself, not least because you cannot measure 'consent' by a majority vote - with so many votes 'for' and others 'against', it measures the very opposite, the degree of dissent. It should also be remembered that the right of self-determination was devised so that colonised peoples could overcome the external problem of foreign domination; it was not intended as a means of resolving internal disputes.
And because you refuse to question this binary basis of the GFA, your concluding para 53 is also confusing if not oxymoronic.
_______________
So why is your Report so reluctant to question the blindingly obvious? Binary voting is ancient, often divisive (as your Report admits), always adversarial, and sometimes hopelessly inaccurate - (as in Brexit). Is it because - I quote your e-mail of 3rd August last - "...it could be very destabilizing to have a vote that was structured such that which option was declared as the winner depended on how the votes were counted." So do you feel it is better if someone(s) decide what the binary choice should be, and the people will then just have to lump it, beef or lamb? AV or FPTP? 'remain' or 'leave'? Were these last two questions not also "very destabilising"?
Let me concentrate on Brexit. If there had been three options, (as I suggested in a press release, four months before the referendum in June 2016) - (a) 'the UK in the EU?' (b) 'in the EEA?' or (c) 'the WTO?' - and if those three options had had the (1st preference) support of 40, 30 and 20% respectively, then potentially, in any binary ballot on a single option, all three could have lost, by 40:50, 30:60 and 20:70 respectively. If the same someone(s) dictated a dichotomy, the outcome could have been (a v b) 40:30, (a v c) 40:20 and (b v c) 30:20, or maybe some folks might have used their 2nd preferences. So, to consider just a little bit of the science, let it be assumed that the 40, 30, 20 had preferences A-B-C, B-A-C and C-B-A.
In binary votes of the 'Option X, yes-or-no?' variety, the outcome would be, as stated, against (a), against (b) and against (c);
In binary votes of the 'Option X or option Y?' variety, the outcome could be (a), (a) or (b).
In two-round voting, the alternative vote, the Modified Borda Count MBC or the Condorcet rule, the outcome would be (b). As I said in my submissions, all four of these methodologies can be used to identify a 'democratic majority opinion' - but your Report ignores them all.
In a nutshell, multi-option voting can be much more accurate than any binary balloting, and preferential voting can be even more precise. Other more complex voters' profiles show that the MBC and/or Condorcet are indeed very accurate, inclusive and robust.
So why does your Report not follow the science, the science of social choice? As you may know, some political scientists have concluded that the most accurate measure of all would be a combined Borda/Condorcet count; indeed, if the MBC winner were to be the same as the Condorcet social choice, everybody could rest assured that this outcome was indeed the will of the people. (Whereas, of course, in the 2011 referendum, because the ballot was binary, some voters were not "free" to choose what they wanted, and so the outcome did not and could not reveal the true nature of public opinion.) If Britain had held three majority votes on the above three options, (a), (b) and (c), then (a), the 'remain' option, would almost certainly have been the winner.
In summary, it may be "very destabilizing to have a vote that was structured such that which option was declared as the winner depended on how" the question was set. Brexit, we now know, was just that, very destabilizing; not only the binary referendum, but also the binary votes of last Christmas Eve in the Commons, they were all hopelessly crude. The 1972 Border Poll was also at least destabilizing. (And the fact that, sometimes, binary referendums can be sort of OK - the GFA poll of 1998, the same-sex marriage referendum in the Republic - should not detract from the fact that, on other occasions, they are at best divisive, at worst provocations to violence - as in the Balkans, the Caucasus, Ukraine, South Sudan, etc..) Hence the above-mentioned de Borda press release of February 2016, in which it was predicted that any binary question would get a negative answer; in its stead a three-option poll was proposed, not unlike the three-option plebiscite which Westminster gave Newfoundland in 1948. Sadly, like my submissions to your Working Group, the press release was also ignored.
So why could there not be an impartial institution - a citizens' assembly or an independent commission - tasked to draw up, say, a five option non-binding ballot - (the number of options which New Zealand's Royal Commission selected for their multi-option two-round referendum in 1992)? Then let the people cast their preferences (which everyone in Ireland is perfectly used to doing). Thus could be identified, at best, the option with the highest average preference; and if that average score were to surpass a given threshold - which varies, depending on the number of options listed - all concerned could know that it also had majority support. Then, by all means, hold a binding binary ratification referendum, so to (again) comply with the GFA. Why could not your Report even mention the fact that this idea had been suggested? Or list the fact that such an idea had been proposed? Or inform me (and not only most or all of the others who had sent in submissions) of the launch? Or apologise on your web-site, as promised, for such unprofessional conduct?
I resort to this open letter because, seemingly, nothing else works.
Yours …
* He was a friend. Shortly before his death, for the work that he had started and which I now continue, he gave me his blessing.