2026-08 Unlock Democracy: egalitarian...
Tuesday, March 31, 2026 EGALITARIAN DECISION-MAKING
Alas, the world still uses the 2,500-year-old majority vote!
On 29th March, yet another international conference ended without an agreement: the WTO’s 14th Ministerial Conference in Yaoundé. Like many another, it worked on the basis of majority voting… as do most democracies, many a theocracy and even, when it suits them, autocracies.
Majority voting and its consequences, majority rule, majoritarianism, and majority coalitions, are ubiquitous and sometimes iniquitous. Today’s majority coalition in Israel is (not the but) a cause of war; majority rule in Northern Ireland led to the Troubles; the word ‘majoritarianism’ translates into Russian as ‘bolshevism,’ большевизм; China also follows this ‘principle,’ 少数 服从 多数 shǎoshù fúcóng duōshù (the minority obeys the majority); and majority voting is prescribed in Article 97 of the Constitution of North Korea – (not that it’s used very often; as per Article 92, parliament meets only once a year).
There are over 300 electoral systems in the world, nearly all of which are regarded as democratic. There are also a dozen or so decision-making systems; of them all, binary voting, the most ancient, probably the most unreliable, is still the basis both of governance with majority rule, and of business with majority shareholdings. Everything else in life has been modernised… but not our decision-making.
MAJORITARIANISM
If disputes are to be resolved in war, people will fight; and most wars are binary.[1] If instead people use (i) a majoritarian democracy, many of the arguments are still binary, so they will argue… and sometimes fight; in contrast, (ii) a preferential procedure could cater for compromise.
(i) In majority voting, it’s “A or B?” (a pairing, as it’s called); or maybe it’s even more simplistic and just “A, yes or no?” (a singleton). Either way, there is rarely any middle ground, and those involved tend to fall into two opposing groups. This primitive, ‘either/or’ methodology is not only adversarial, it is also often inaccurate.
Variations involve simple, weighted or consociational majorities, but in all these formats, the adversarial nature sill dominates.
(ii) Multi-optional and preferential voting offer three or more options. At its simplest, with three options – let’s say, A, B and C – the voters may choose singles, pairs or triplets, depending on the voting procedure. They might vote for just one option – A, B or C – or cast two preferences – A-B, A-C, B-A, B-C, C-A or C-B – or list all three – A-B-C, A-C-B, B-A-C, B-C-A, C-A-B or C-B-A. With four options, there are 24 ways of submitting a full ballot. With five, 120, whence we may relish the diversity of our species… but still get a definite outcome. Pluralism is possible.
Suffice here to say that with a points form of preferential voting, folks don’t vote ‘for’ or ‘against’, but ‘with’ each other: you state your preferences, I state mine, but nobody votes ‘no’ to anything, nobody votes ‘against’ any thing or any body. Thus can we identify, not the more popular of just two options, but the most preferred of several. More of all this in a moment.
VOTING PROCEDURES
For nearly 2,000 years, many clever minds have realised that, when there are three or more options ‘on the table,’ binary voting has its limitations. Hence, various forms of multi-option voting; they include the following.
+ Plurality voting dates from the year 105 in Rome; the voter may cast only one preference (which, with tactical voting, may or may not be her 1st preference), and the winning option may enjoy, not majority support, but only that of the largest minority. It was first used in governance, not in Europe but in China in 1197, during the Jīn Dynasty.
+ (We fast forward to 1790, when) the French devised the two-round system TRS – a plurality vote followed perhaps by a majority vote – which ensures the outcome has majority support, perhaps, and gives the voter a little more room for manoeuvre.
+ Resuming a chronological order, approval voting was first used in Venice in 1268; there are no preferences, the voter ‘ticks’ the options he likes/accepts/tolerates, and the option with the most ticks wins. At worst – if everyone ticks only one option – it is little better than a plurality vote.
The next four methodologies are preferential.
+ The Borda Count BC of 1433 is a points system: in an n-option poll, a 1st preference gets n points, a 2nd gets (n-1), and so on; {that or (n-1), (n-2) etc.}; and the option with the most points wins. It might be the option with very few 1st but lots of 2nd preferences. This BC can be non-majoritarian but, like approval voting, it might not be.
+ In 1770, Jean-Charles de Borda proposed the Modified Borda Count MBC: if, in an n-option poll, a voter casts only m preferences, the 1st preference gets only m points, the 2nd preference gets (m-1), and so on. So, in a five-option ballot, she who casts only a 1st preference gets her favourite only 1 point; while he who casts all five preferences gets his favourite 5 points, his 2nd choice 4, and so on. At best – if (nearly) every voter casts a (nearly) full ballot – the MBC can identify the option with the highest average preference; and an average, of course, involves every voter, not just a majority of them! The MBC is inclusive, literally! It is non-majoritarian, always.
+ In 1795, Le Marquis de Condorcet proposed a rule which analyses all the pairings; with three/four/five options, as implied above, there are 6/12/20 pairings – (and even more different full ballots, 6/24/120) – and the Condorcet (or Copeland) winner is the option which wins all (or most) of them.
+ Thomas Hill’s alternative vote, AV, of 1821, is also known as the single transferable vote STV or ranked choice voting RCV; it is another knock-out system, a series of plurality votes. Each round eliminates the least popular option and transfers its votes according to its 2nd or subsequent preferences, until the winning option has a majority. It’s a very good system, but can be capricious; what’s more, the option which is the 1st preference of none but the 2nd of all – the perfect compromise? the most democratic outcome? – may get a stage (i) score of zero and be eliminated.
In all of the above procedures, a majority opinion can guarantee a win with but one exception; uniquely, as said above, the MBC is always non-majoritarian. And, after all, democracy is for everybody, not just a/the majority, the rich, the poor, the Protestants, the Hutu, the Serbs, the Jews, the Sunni etc., etc.; it is for everybody!
Politics is the art of compromise, and the MBC is its science; to be democratic, a voting procedure should be egalitarian.
ACCURACY AND DEMOCRACY
The above procedural differences can be consequential. Plurality voting excludes any voters who might otherwise support a compromise: in all probability, such an option is not even on the ballot paper.
TRS can be a little better, especially if the voter can guess how others are intending to vote.
And AV is definitely better: it enables the voters to support a compromise; experience (of its use in an electoral system, PR-STV) from Northern Ireland shows, however, that while the more moderate voters often use this freedom of choice to cross the gender gap, the party divide and even the sectarian chasm, the more extreme voters tend to remain entrenched.
The MBC goes one stage further: it encourages the voters to be inclusive. If you cast only one preference, your favourite gets only 1 point; but if you cast n preferences – if m = n – your 1st preference gets n points, your 2nd gets (n-1) and so on. In other words, if you want your favourite to win, it’s better to cast your compromise preferences as well, to vote not ‘against’ but ‘with’ your neighbours.
Exclude, enable or encourage distinguish the three; but a democratic voting procedure should not prevent, it should at least permit or at best promote more participation.
Of the methodologies listed, only the MBC and the Condorcet rule take all the preferences cast by all the voters into account, always! They are, it is argued, the most accurate, and can be compared to a sports league. Every team (option) plays (is compared with) every other team, and the team (option) which wins the most matches (pairings) is the champion (Condorcet winner). In most seasons, this victor has a very good if not the best goal difference (points). In other words, the Condorcet winner is often (but not always) the MBC social choice as well. Both procedures are very accurate; little wonder, they often produce exactly the same results.
THE THEORY
As noted, politicians argue, partly because they use majority voting which is indeed adversarial. Consider now what could, and what would probably happen in an assembly or parliament, if the final decision-making process were subject to a five-option MBC ballot.
To win, the protagonist of Party X, in moving option X, may need not only lots of 1st preferences (from the Party X supporters), but also several 2nd preferences from other voters, and maybe a handful of 3rd’s… but few if any 4th or 5thpreferences! Furthermore, Party X will hope its supporters cast full or nearly full ballots, to state many if not all of their preferences, so that its voters’ 1st preferences get a full 5 points each rather than anything less. Every protagonist should therefore best declare his/her own preferences, in order to:
(a) persuade its party colleagues to cast full or nearly full ballots,[2]
and
(b) woo those of other parties to cast, as mentioned, a 2nd or at least a 3rd preference in support of option X.
If (nearly) every party does act in this way, the ambiance of the debate will be more constructive, and the result very transparent.
THE DEBATE
In any debate in a multi-party parliament, every party shall be eligible to propose one option, and every option which does not contravene the UN Charter on Human Rights shall be allowed ‘on the table.’ In a five-party House, therefore, debates may well involve five options. And probably what would happen in chambers of many, let’s say, ten parties, is that two or more smaller parties might join forces, or a small party may align itself with a larger one. In most democracies, most debates will usually consider about half-a-dozen options.
Debates may be complex, in which case the Speaker may choose to divide the debate into various sections. The relevant proposals of every party shall be debated in turn; each option or clause may be amended, composited or even deleted… but only if the proposing party agrees to such a change. The number of options ‘on the table’ may therefore vary.
Now ideally, and especially if the debate is complex and/or controversial, an MBC should consist of 4 – 6 options. If the number of options increases, the Speaker may propose either a short list of up to six options, or more than one ballot.
THE BALLOT
A binary vote may be used if, and only if, its dichotomy is a duality. “Shall we drive on the left or on the right?” is a good example, for there are no other options.[3]
If and when there are more than two options ‘on the table’ – which, in any debate on a controversy in a pluralist democracy, will be almost always – the Speaker shall draft one or more ballots, each of about five options, such that the (short) list of options is balanced and covers, as best it may, the full debate.
Prior to the vote, the participants shall first be allowed to comment on the draft.
THE VOTE
As stated, the MBC encourages the voter to cast, not only a 1st preference, but also a 2nd and subsequent preferences; in effect, to state her favourite option, but also any compromise option(s). In so doing, she recognises the validity of the other parties’ options which, by virtue of their presence on the ballot, the Speaker has already deemed to be valid.
If (nearly) every voter does cast a (nearly) full ballot – in other words, if every voter does state their compromise option(s) – the most popular option will represent the collective compromise… which is what democracy is meant to be all about. The count must nevertheless cater for those ballots in which some choose to submit only partial ballots, so instead of comparing average preference scores, we use consensus coefficients, CC.
An option’s CC is defined as its MBC score divided by the maximum possible score. In a five-option debate, this may vary from the maximum of 1.00 to a minimum (if everyone has cast a full ballot) of 0.20; and therefore the CCMEAN will be 0.60. An unlikely result could be a three-/four- or five-option ‘dead heat’… but in that case, with no obvious most popular option, it would be obvious that there was no consensus and so, at a later time, the Speaker may call for the debate to be resumed. In many instances, however, something will be seen to be more popular than the other options.
The following figures are somewhat arbitrary; the Speaker may wish to adjust them, as and when the participants become more used to consensus decision-making. But if:
0.65 < CCWINNER < 0.75
the result may be the best possible compromise.
If instead
0.75 < CCWINNER < 0.85
it may deserve the term, ‘consensus’.
And if
0.85 < CCWINNER < 1.00
it could be called the collective wisdom.
IMPLEMENTATION
Having identified ‘the will of the people’ or at least ‘the will of their representatives’ (which, if the electoral system is fair, should be similar), the democratically elected members of the House shall then implement it, regardless of whether or not the outcome was their 1st, 2nd or 3rd … or even their 4th or 5th preference. As democrats, they should work to ensure the democratic will, once identified, is then enacted. If for whatever reason they feel they cannot, they can but resign.
COP31, and FUTURE COPs.
Democracy is often adversarial. It need not be so. If the final decision-making process were to be less divisive, or at best consensual, the procedure could encourage not confrontation but cooperation.
Some of the very few organisations to reject majority voting are the UNFCCC’s annual COP meetings.[4] For reasons unclear and unstated, the UN has not yet considered preferential voting; instead, in what often become protracted, late-night debates, the parties seek a verbal consensus while allowing participating nations to apply a veto – the very opposite of consensus!
Sadly, many nations in the UN – and many politicians in national parliaments – want either to win everything or to lose nothing. With the MBC, however, no one party wins everything, and (nearly) every party wins something. Uniquely, the MBC is egalitarian. It is ideally suited to such large international gatherings, devoted as (they say) they are to an egalitarian world.
Such a non-divisive procedure could also be used to good effect in divided conflict zones, and in national parliaments, many of which, at the moment, divide… unless, they say, like the UK in WWII, there is an emergency. But today’s world is already confronted by at least two existential emergencies: climate change, and wars.
CONCLUSION
An egalitarian future will need an egalitarian voting procedure of decision-making.
______________
Dr. Peter Emerson. (A Russian speaker and student of Chinese; he has worked in Moscow, Kiev, Sarajevo and Tbilisi, as well as in Beijing, Hong Kong and Taiwan.)
[1] When Finland fought the anti-fascist USSR in WWII, a binary analysis was too simplistic. Similarly, today in the Middle East, with one war between Iran and the USA coinciding with other conflicts – Israel v Palestine, Sunni v Shi’a – the mix of several binary battles is confusing, horrible and dangerous.
[2] To recognize the validity of all but the most extreme parties’ options. Parties in Germany, for example, may choose to list all the proposals, except that of the AfD.
[3] Only one country has every held a referendum on this topic – Sweden in 1955 – with not two but three options on the ballot paper: ‘left’, ‘right’ and ‘blank’. But ‘blank’ enabled those who were indifferent to still participate in the democratic process; it wasn’t a third, independent option.
[4] United Nations Framework Convention on Climate Change; and Conference of the Parties. In November 2024, the author attended COP29 in Baku.
