What is “Majority Rule” and is it a good idea?

A rumination on why I think “democracy” has to mean more than “majority rules” or “the favorite wins”—even when only a single candidate or proposal is being chosen.

The possibility of Condorcet “cycles” infecting the preference-rankings of groups is pretty well known by now—especially since Arrow’s impossibility theorem. The idea is that a group entirely composed of individuals whose preference-rankings are transitive may end up liking (as a group) A more than B, B more than C, and C more than A. This can happen because different sub-groups make up the three aggregate ratings. This (and other voting paradoxes even involving pairwise comparisons and Borda counts) have led some observers to denounce majoritarianism. Such critics consider it either an approach that can’t provide unambiguous winners when there are more than two choices, or worse, something that unambiguously provides the wrong answer.

Now, as I look at these matters, there are at least two essential characteristics of fair democratic choosings. First, they are egalitarian in this way: they must, to use the old Benthamite language, “count each vote as one and none as more than one.” That is, they cannot countenance weightings of most kinds, whether they are considered to follow from any rankings (cardinal or ordinal) of the voters or from any external assessments regarding the value of this or that vote or voter. Second, they are egalitarian in another way: the authority granted winners of elections must, in some rational manner, reflect ratios involving both the number of eligible voters and number of votes received. (I will not take up this latter requirement in this OP.)

While simple majoritarianism seems to share both of those desiderata, I take it that the latter (my own view) can’t rightly be characterized as a majoritarian position itself because it does not accept what is commonly known as “the majority criterion.” What is that? It simply requires that If there exists a majority that ranks a single candidate higher than all other candidates, that highest-rated candidate must win. As will be seen, there are good reasons for those with sound democratic principles not to join with majoritarians on this matter. In any case, the (let’s call it) “Egalitarian Proportional Democracy” I’m pushing for here shares with majoritarians the views that political actions and offices must be taken and distributed on the basis of the number of voters who want or don’t want something, rather than on how much they want them (as well as on the other matter that I’m not planning to discuss here). But surely that doesn’t tell us very much. Can at least the egalitarian portion of my description of Egalitarian Proportional Democracy be fleshed out? Let me try.

Suppose eight people are having a party and are trying to decide what soda to bring. [Based on an FMM comment, I add here the assumption that, for whatever reason, it would be a major hassle for there to be more than one choice of soda at the party.] And let there be four possible choices: Cola, Lemon-Lime, Orange and Root Beer. There’s no unanimity among the revelers, so, being the good (small-d) democrats they are, they think that the majority ought to have its way and plan a vote to decide the matter. Here is the result when they are asked to give their favorite (here designated with ‘X’):

          A  B  C  D  E  F  G  H
Cola   X  X  X
L-L                  X  X
Orange                     X  X
RB                                    X

As can be seen, while Cola receives a plurality of the vote, no flavor gets a majority. One member therefore suggests a run-off with the first and tied-for-second contenders only, leaving off RB all together since it did so poorly. Here are the results of this run-off election (with ‘A’ indicating an abstention):

           A  B  C  D  E  F  G  H

C         X  X  X                    A
L-L                    X  X            A
O                               X  X   A

This vote didn’t help: there has been no movement at all because voter H absolutely loathes all the flavors except RB and refuses to pick any of the others as a passable choice for the party.

The revelers aren’t completely stuck though, because there are other voting possibilities. Let us suppose that, like me, this group has no truck whatever with the inter-personal assessments of preference intensities required for cardinal ordering, and that they are also skeptical of ordinal rankings to the extent that those assume similar “distances” between preferences. They think, that is, that there could be a huge divide between one person’s 1st and 2nd choices, and hardly any at all between another ranker’s 1st and 2nd picks.

Fortunately, two members of this group have been regularly assaulted by emails from voting reform organizations: one, from a group that pushes Approval Voting (“AV”), and another that favors Score Voting (“SC”). Those two discuss the matter with the other six party planners and the SC advocate is able to convince everyone that they can exclude all the questionable preference weights by using the following scale:

FAVORITE……………………………………………………………………4 PTS
GOOD ENOUGH (WOULD DRINK IT IF AVAILABLE)………………….3 PTS
PASSABLE (NEVER HAD BUT WD TRY IT IN A PINCH)……………..2 PTS
NOT OK (NEVER HAD & WON’T TRY EVEN IF THIRSTY)…………..1 PTS
REALLY DISLIKE………………………………………………………….. 0 PTS

The AV supporter is on board with undertaking a new vote that would use this scale, but only if the assignments of 4, 3, or 2 points are counted as “Approvals”—meaning that the voter can “live with” the choice. This is agreed upon as well, and the third vote is taken. For ease of counting, I represent the approvals here with an “(A)”:

           A        B      C       D       E       F       G       H     TOT.     Apps
C      4(A)   4(A)   4(A)   2(A)   2(A)   1        0        0      17        5
L-L     2(A)   2(A)   2(A)  4(A)   4(A)  2(A)  2(A)     0      18        7
O       3(A)   2(A)     0     3(A)   3(A)  4(A)  4(A)     0       19       6
RB     3(A)    0        0      0        1      3(A)  2(A)     4(A)   13       4

As can be seen, while the Plurality victor was Cola, the SC winner is Orange and the AV winner is L-L!

Perhaps it will seem that this embarrassment of “winners” is the result of the weirdness of there being so many “never tried it” votes with respect to what seem like common carbonated drinks. But it is important to realize that an attitude of “I really don’t know much about her (or it).…” toward political a political candidate or proposal isn’t unusual at all. Look at the results above again, but this time, think of it as a political election for a representative, with each coming from a different Party. (Perhaps replace “Cola” with “Corporatist”; “L-L” with “Liberal”; “Orange” with “Outsider” and “RB” with “Republican”.) This may make it clearer that there can be a large number of decisions in which the assignment of one or two points (approval or disapproval) will largely be a function of the varying amounts of risk that voters are willing to take. Some people will be OK with this or that relatively unknown candidate or proposal; others will not be willing to take any chances.

Keeping all this in mind, which “winner” will the authentic egalitarian support in this election? The Corporatist, because he is the favorite of the largest number of voters? The Outsider, who got the highest score? Or the Liberal, who most voters found to be minimally acceptable? In my view it is the number of approving voters that the sensible democrat must take to matter most. Just as we ought not to be stuck at parties with nothing we can stand to drink, we ought not to be stuck with ruler/representative A when more people among us can stand candidate B. On this view, if it is to be used to determine what “the people” do or don’t want, majoritarian/egalitarian-style aggregation should be understood as the counting of approvals, where each person’s approval is given the same weight as everyone else’s, regardless of how enthusiastic or tepid it is. That tack definitely seems more conducive to stable regimes than one in which candidates that a ton of the populace don’t approve of get to take office.

That is my current take on the matter. I recognize that I have here avoided all of the complicated issues surrounding strategic voting and how that is likely to affect results (if you’re curious, see the Wikipedia article on “Approval Voting.”) Anyhow, I look forward to comments to get a better handle on this. Thanks.

282 thoughts on “What is “Majority Rule” and is it a good idea?

  1. timothya:
    Instant recall means that policy and practice is set directly by the electors, with open debate and in the clear light of day. Are you suggesting that ordinary people cannot be trusted to making good decision? Or are you suggesting that good policy is better set by lobbyists, special-interest groups and The Faceless Men of political hackery? I know which model I support, and history supports me.

    Either you live in never-never land, or you don’t really understand how your system works, only how you are told it works. Sure, ordinary people can be trusted to make good decisions. Of course, they need to be fully informed, which few generally are. And those who aren’t well informed somehow need to realize that the informed people are making good decisions. And in practice, a dozen well informed people will come up with at least thirteen ideas, and nobody will know which is best (and there might be better ideas than any of these).

    So how about a system where the people select from among their number those they consider best qualified to make such decisions, and require that their deliberations be open and published? Of course, for a large and varied population with over a million different perceived self-interests, any workable program will require attention to over a million details in implementation and in adjudication — and there will always need to be adjudication. And this in turn requires quite a large body of involved people to write the details, adjudicate the disputes, enforce the policies and decisions. This requires a bureaucracy, whatever you wish to call it.

    I must say, I do envy those who live in countries where opinions are unanimous, where details don’t matter, where everyone is happy with every decision. Failing that, I’d love it if nobody was corrupt, if there was no self-dealing, nothing happening out of sight. I don’t believe your system achieved this.

  2. Alan Fox:
    walto,

    Representative in an ethnic and gender sense. Not all old white guys.

    Ah, so you mean representative a la identity politics.

    Two in their 20s, one trans, two black, two gay, one ginger. More Irish, more Latino, fewer Jews. Got it.

  3. Flint: I’d establish reasonable term limits for all legislators in Congress and the states — and for judges, who would NOT be electec. I’d establish tight limits for gerrymandering along political as well as racial and religious lines. I’d eliminate voter registration altogether and take steps only to ensure that people only voted once. I’d declare election day a national paid holiday. I’d prohibit lobbyists from writing legislation. I’d establish progressive taxation without loopholes so that 0.1% of the population wouldn’t be sucking up 50% of the wealth, and use that money to subsidize good education — especially about how democracy works and how it can be corrupted. I’d prohibit private campaign donations, and use taxpayer money to distribute fairly to all candidates who meet some minimum criteria (say a minimum poll percentage – and the candidates wouldn’t be doing the polling).

    And I’d probably never get elected, or get ANY of this through Congress. Not without a generation of bloodshed.

    Well, I’d definitely vote for you. Although you would have to get that (rather interesting) “no voter registration” idea passed first (heh), and that’s gonna drive the nativists insane
    I do worry that term limits may produce a transfer of power to ‘professional’ civil servants, a la Yes Minister. I think that your campaign finance reform would get you most of the way there…
    P.S. I also liked the ingeniousness of the idea (was it yours?) of allowing all interested parties to submit their redistricting plans, and the one with the shortest borders wins. I’d like to include some homogeneity metrics too, but that’s where the arguments start…

  4. walto: How can something be apolitical and representative? Who/what is or isn’t representative is a political matter.

    Exactly. Elect the Supreme Court. For fixed terms.

  5. timothya: Exactly. Elect the Supreme Court. For fixed terms.

    Can anyone cite the reasons for the way things are? I mean, there are records of the thinking of the people who wrote the American Constitution.

    It would seem useful to recapitulate their debates and to re-argue the issues.

    More so than argumentum ad butthurt.

  6. petrushka: It would seem useful to recapitulate their debates and to re-argue the issues.

    There are, among many many examples, the Madison notes, the Beard book on the Constitution, and the Garry Wills book. Go get ’em and report back.

  7. Flint: Either you live in never-never land, or you don’t really understand how your system works, only how you are told it works. Sure, ordinary people can be trusted to make good decisions. Of course, they need to be fully informed, which few generally are. And those who aren’t well informed somehow need to realize that the informed people are making good decisions. And in practice, a dozen well informed people will come up with at least thirteen ideas, and nobody will know which is best (and there might be better ideas than any of these).

    So how about a system where the people select from among their number those they consider best qualified to make such decisions, and require that their deliberations be open and published? Of course, for a large and varied population with over a million different perceived self-interests, any workable program will require attention to over a million details in implementation and in adjudication — and there will always need to be adjudication. And this in turn requires quite a large body of involved people to write the details, adjudicate the disputes, enforce the policies and decisions. This requires a bureaucracy, whatever you wish to call it.

    I must say, I do envy those who live in countries where opinions are unanimous, where details don’t matter, where everyone is happy with every decision. Failing that, I’d love it if nobody was corrupt, if there was no self-dealing, nothing happening out of sight. I don’t believe your system achieved this.

    Who is talking about unanimity? Debate, vote and the majority rules. Simple and effective.

    And for all its failings (most of which derive from a refusal to govern the behaviour of plutocratic organisations), my country’s government process is reasonably effective:

    1. The parliament passes legislation (embodying the purpose and outline of the act and identifying the body responsible for implementation)
    2. The ruling party writes the Regulations, which specify how the legislation will work in practice
    3. The parliament votes money supply for the implementation (or does not, in which case the legislation lies void)
    4. The bureaucracy implements the regulations up to the limit of the budget
    5. The relevant minister is accountable for performance and must report to Parliament

    I speak from experience in saying that these roles and responsibilities are rigidly enforced.

  8. DNA_Jock:
    P.S. I also liked the ingeniousness of the idea (was it yours?) of allowing all interested parties to submit their redistricting plans, and the one with the shortest borders wins. I’d like to include some homogeneity metrics too, but that’s where the arguments start…

    The shortest border districting idea was mentioned here by petrushka, but it has been kicking around the literature for decades. Computer-drawn district boundaries are certainly an improvement over intentional gerrymanders, but they don’t serve much purpose beyond that, and can cause harms of their own by splitting cities, counties, ethnic communities, etc., Since there’s no reason to suppose that geography is the sole (or even the best) way of getting minority representation anyhow, computer-drawn boundaries mostly just make a bad concept a little better by preventing it from being made much worse.

  9. We use the preferential voting system (Im from Oz).
    Its bloody complicated but seems to be logical/fair.

    If there was an obviously best system, everyone would be using it. Well, everyone would know about it.

  10. graham2: If there was an obviously best system, everyone would be using it.

    No, that’s a mistake. Some people prefer a system where they can have undue influence on the results.

  11. graham2: If there was an obviously best system, everyone would be using it. Well, everyone would know about it.

    Dunno about “obviously,” but as to “best,” some systems are better for producing some sorts of outcomes, some are better for producing other sorts. Some are cheaper to run, some provide more voter information, some are easier to understand, some achieve better turnout, some are harder to game, etc. So it depends on your goals.

    In my own view, it would be best to have two distinct voting systems operating in a connected way. No single system can do everything that ought to be done. As indicated above, I personally like AV for finding a single winner in any race.But that’s not all a democratic system should do, I don’t think.

  12. Flint: Either you live in never-never land, or you don’t really understand how your system works, only how you are told it works. Sure, ordinary people can be trusted to make good decisions. Of course, they need to be fully informed, which few generally are. And those who aren’t well informed somehow need to realize that the informed people are making good decisions. And in practice, a dozen well informed people will come up with at least thirteen ideas, and nobody will know which is best (and there might be better ideas than any of these).

    FWIW, I don’t agree with too much of this paragraph. Specifically, I don’t think voters need to be well-informed or anything like that. I mean, I may want them to be, but that’s neither here nor there, IMO.

    So how about a system where the people select from among their number those they consider best qualified to make such decisions, and require that their deliberations be open and published? Of course, for a large and varied population with over a million different perceived self-interests, any workable program will require attention to over a million details in implementation and in adjudication — and there will always need to be adjudication. And this in turn requires quite a large body of involved people to write the details, adjudicate the disputes, enforce the policies and decisions. This requires a bureaucracy, whatever you wish to call it.

    I must say, I do envy those who live in countries where opinions are unanimous, where details don’t matter, where everyone is happy with every decision. Failing that, I’d love it if nobody was corrupt, if there was no self-dealing, nothing happening out of sight. I don’t believe your system achieved this.

    I like all the rest though.

  13. walto: And also I don’t get the connection between my OP and that study. Can you explain?

    Sure, The study shows that small groups have a very different dynamic than larger groups. They avoid many of the problems with “majority rules” that your approach is meant to remedy.

    That is why federalism is the solution.

    peace

  14. The OP didn’t mention federalism, which is orthagonal to it–you brought it up in spite of it being a bit off-topic. Just another pet something you wanted to thump here. And anyhow, the study you linked is, well….here’s the conclusion:

    “Historically, people have looked at social networks as equivalent to an anatomical network based on static ties between people. We are putting forward the idea that in small groups, networks evolve in time based on actions,” he said, adding that these time-based actions are like networks in the brain, where physically distant neurons forge connections toward a specific function. “Our approach is analogous to learning about neural circuits based on how they function in the brain, rather than how they are anatomically connected.”

  15. walto: The OP didn’t mention federalism

    I know it brought up supposed problems with majority rule and your proposed solution.

    Federalism is just a much better solution as witnessed by the study .

    There is a reason why neurons work the way they do and a reason why mimicking them if possible is a good idea.

    peace

  16. fifthmonarchyman: Federalism is just a much better solution as witnessed by the study .

    Solution to what problem exactly? Each sub-unit has to decide stuff too. And then there needs to be a way to combine those decisions federally. As I said, entirely orthogonal, in spite of it being one more of your thangs.

  17. walto: Each sub-unit has to decide stuff too. And then there needs to be a way to combine those decisions federally.

    Sub groups should be small enough so that decisions are made by consensus.

    Ideally every step up the hierarchy should decide matters in that same way. The sub groups chosen representatives should also decide by consensus.

    There is a reason why the senate is supposed to be a deliberative body and why each state has equal representation. 50 states make careful deliberation more difficult than 13

    When a group becomes too big we should look at breaking it up or moving it’s power down to a level where consensus making is still possible.

    Larger groups are always in danger of mob rule and using factional politics to screw the minority.

    That’s my take anyway.

    It’s a power problem and not a voting problem.

    peace

  18. As has already been discussed above, with a consensus-only system, one person who doesn’t want to do anything will always get to run the show. Not just “the minority” gets screwed, the majority does too.

    Anyhow, that picture has been discussed to death and thoroughly debunked since Hobbes and Locke pointed its inconsistency with actual democracy a looong time ago. Still have abuses arising from it on certain types of juries though. One person gets to run it, if he/she has enough endurance. In societies, it’s generally wealth that’s the more important factor, since the wealthy don’t usually need the group to do anything at all.

  19. walto: one person who doesn’t want to do anything will always get to run the show.

    That is what excommunication is for. If no consensuses can be reached then you have two groups rather than one.

    walto: Still have abuses arising from it on certain types of juries though. One person gets to run it, if he/she has enough endurance.

    When it comes to a Jury you end up with a mistrial if consensus can’t be reached rather than one person running the show.

    Like I said before federalism is not perfect but it is the best system AFAIK.

    peace

  20. fifthmonarchyman: That is what excommunication is for. If no consensuses can be reached then you have two groups rather than one.

    That’s how Carl Schmitt saw it. Those who disagree are enemies of the people. Spinoza got the boot too, as I recall.

  21. And let me ask you this. 50 people can’t agree on whether to fix up your church, tear it down, or do nothing. The split is 35, 10 and 5. They debate and debate, no progress. Do the 5 (do nothing) guys win, or do you excommunicate them?

  22. walto: In societies, it’s generally wealth that’s the more important factor, since the wealthy don’t usually need the group to do anything at all.

    Wealth is only a benift if the rest of society values what you have.

    A young man once told Confucius that if he kissed the butt of the emperor he would not have to eat lentils all the time.

    Confucius replied that if the young man learned to enjoy lentils he would not have to kiss the emperor’s butt all the time

    peace

  23. Lentils, got it. Or they could eat cake, presumably.

    Anyhow, I have another question for the group. What should the voting age be?
    6?
    12?
    16?
    18?
    21? or
    25?

    Thank you all in advance.

  24. walto: They debate and debate, no progress. Do the 5 (do nothing) guys win, or do you excommunicate them?

    Usually they get tired and concede or move on. At least that is what should happen. What actually happens depends on the circumstances

    Often folks just smile and ignore them like you do with a crazy old uncle. 😉

    peace

  25. Neil Rickert: No, that’s a mistake.Some people prefer a system where they can have undue influence on the results.

    Another mistake. Some people have undue influence (the Dear Respected, for example), and prefer a system that makes the results irrelevant.

  26. fifthmonarchyman: Often folks just smile and ignore them like you do with a crazy old uncle

    Would you ignore them by going ahead with one of the other two proposals (in which case there would have been no consensus and you’d need another system for determining a winner); or would you ignore them by just forgetting the whole rebuild thing (in which case the smallest group of the bunch would have gotten their way)?

  27. J-Mac: No offense to walto but I’m confused why this OP is published at TSZ, and even more, why it is featured…

    Democracy is a complex of interrelationships integrating many different levels of organization of a type that are often found in biology, including multicellularity, neural networks, and ecosystems.

    fifthmonarchyman: I think that when a body gets too big for effective consensus building the solution is to split it up into smaller parts.

    There are reasons for centralization, such as defense against centralized threats (e.g. fascism), and to end destabilizing threats within smaller units (e.g. slavery in the U.S.). That means there are countervailing currents between low levels of organization and high levels of organization. What happens is a complex of interrelationships develop that pass information and power between the different levels.

    fifthmonarchyman: The solution in my opinion is to devolve power more localy so that folks in Kansas are not so scared about what the government in far off DC can do to them. That approach might have worked with Hitler as well.

    The Weimar Republic had a {federated} diffusion of power, but the institutions weren’t well-established. When Hitler came to power, he consolidated power {including ending federalization} by assassinations, arrests, and other means. Many people on the right stood by while it happened because they wanted to break the influence of the left.

    fifthmonarchyman: All that is necessary is for someone on the Supreme Court to begin to take the 10th amendment seriously again and strike down federal laws that overstep constitutional bounds.

    The Supreme Court has limited federal laws based on the Tenth Amendment with the Anti-Commandeering Doctrine, including recently, the Medicaid expansion of ObamaCare. (See National Federation of Independent Business v. Sebelius.)

  28. Flint’s entire comment is worth a read.

    Flint: Just in general terms, an effective government requires some meaningful feedback from those who are governed.

    This hones in on the reason why democratic societies have been generally more successful than autocratic governments. The feedback takes time, but over the long run, leads to changes that make the central government more responsive, as well as more stable. More particularly, power is exerted and shared at many levels, and information is diffused throughout the system.

    Flint:someone loses in every court decision, and we rely on the losers to accept the verdicts. Someone loses every election, but is willing to work with the winner if the election was fair.

    Majority rule is arbitrary! This is especially clear when the difference of who wins an election to the powerful position of U.S. president is only a few hundred of a hundred million votes, as in 2000.

    But while the rule is arbitrary, it is the willingness of the minority losers to accept the results knowing 1) their liberty will be protected, including the liberty to try again next election; 2) there was no undue corruption in the process. When the elected leader rules as if he represents both the winners and losers, then people are more likely to accept the process. But when the elected leader sets about to punish his opponents and reward his followers, and especially when new elections can’t be called immediately, then confidence in the system is degraded, and instability ensues.

    Flint: careful structure and sensible rules, faithfully applied, can make things easier.

    The ideal is accountability, where the best aspects of people are encouraged, while the worst are discouraged. In the U.S., this might entail the Congress protecting its prerogatives against the onslaught of executive overreach, but the current exaggerated party partisanship means most politicians are beholden to their party, not to the institution.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.