Shmuel Gerber

Yes, I think J.J. has illustrated my point. The minority protected is any greater than 1/5. This is equivalent to 24/120. So with 120 voters in total, a minority of more than 24, by coordinating their votes in a cumulative-voting election for 4 seats, can be sure to elect one candidate.

As I said in the topic I referred to earlier, I don't really cotton to the notion that there is something special about a majority of the ballots cast in an election being done by cumulative voting. It does not represent a majority of the voters or of the votes cast. It is just some strange artifact of the rules relating to the way a ballot vote is conducted when a majority of the ballots cast represents a majority of the voters. However, simply for the purpose of this discussion, I'm willing to assume that in order to be among the winners, a candidate must not only be in the top 4 vote-getters (where there are 4 positions) but also must attain a number of "votes" (in the cumulative system, i.e., where a voter can assign multiple votes to a single candidate) that is a majority of the number of ballots cast. As I believe J.J. pointed out earlier, this means more than 60 votes, which then translates to 61 when dealing with whole numbers. I'm beginning to see where you're getting these numbers from — i.e., each voter can place 4 votes, so for a candidate to attain more than 1/2 the total ballots, the minimum number of voters who could yield this majority is more than 1/8 of the voters, all giving their total support (⅛ × 4 = ½). I don't think it's helpful to convert this into a particular number of votes, as it stems from a fraction that always applies, regardless of how it may translate into whole numbers under specific numbers of voters; but, yes, in the case of 120 members voting, 16 voters (all giving all of their votes to the same candidate) would be needed for a majority. But I don't agree that this is the number (or fraction) that is protected by the rules. As J.J. has noted, such a majority does not necessarily win an election even when cumulative voting is in effect. In fact, if a certain set of 4 candidates is supported by more than 4/5 of the voters, and a different candidate is supported by a minority of more than 1/8 (but not 1/5 — which cannot exist in this scenario, because the total minority remaining is not that large), then the minority candidate is sure to lose (assuming all the votes are well coordinated). I would think that if anything, where a majority vote is required with cumulative voting, the rules protect not the minority of more than 1/8, but rather the majority of 7/8, by ensuring that 1/8 or less cannot elect their choice (without additional support). But even that protection is not really necessary, because 4/5 (which is even less than 7/8) can get what they want regardless of the "majority" threshold. I still say that the minority that is protected by the rules is a minority of more than 1/5, which can always (with proper coordination) elect one candidate of its choice in a 4-seat election.

Five just happens to be the number of candidates in JJ's example; there could have been more. There are still only four seats being elected.

I would say yes, but apparently there's still some confusion (and I do not understand why) about what "total votes" means and how it affects the formula.

I don't understand this at all. When we say "four-fifths vote," we are talking about four-fifths of the votes cast. That is the fraction which remains constant regardless of the number of votes cast.

Not exactly 20%, but any number greater than 20%.

Suppose an election rule said directly that a 4/5 vote is required for a candidate to attain the position. Would you still say that something other than a four-fifths vote is needed to suspend the rules, because the election may be decided by one vote? Or that we cannot know what vote is needed because we won't know how many votes will be cast in the election until the actual vote on the election is taken? It seems that there are (at least) two separate discussions going on here. Dan Honemann has already agreed that if the fraction of votes necessary remains constant, then the vote to suspend the rules is dependent on that fraction regardless of how many actually vote. The way I see it, the rule in this case (cumulative voting that started the thread) protects a minority of greater than one fifth. If you want to argue that even if the fraction remains constant, there's no definite minority because the number changes depending on how many vote, that is another discussion.

I don't understand any of these complicated calculations or where they're coming from. In an election of four committee members in which cumulative voting is allowed, a minority of more than 1/5 (that is, more than 1/(4+1), or 20% of the votes cast) is guaranteed election of one candidate of their choice if they put all of their votes on that choice. Therefore, it seems to me, suspending the rule permitting cumulative voting where there are four winners to be elected cannot be done by less than a 4/5 vote. As far as where a majority enters the calculation, I don't think a majority vote is generally needed for cumulative voting, but Mr Martin has opined here that if it is, what is required is for a candidate to obtain a number of votes that is more than half of the number of ballots cast. Here, if more than 1/5 of the voters cast each of their four votes for a single candidate, that candidate's votes will exceed 4/5 of the number of ballots cast, which is easily a majority. Now if there are some other calculations (which I don't understand) that show that even 20% or less can somehow be sure to elect a candidate depending on how many votes are cast, that certainly should not mean that the rule can be suspended by a 2/3 vote rather than a 4/5 vote! If someone can tell me where I've gone wrong, I would appreciate it. Although I would prefer to be told that I am absolutely correct. 🙂

I was not able to follow that. You gave several numbers depending on "my scenario" (Dan) or "your scenario" J. J., but I don't see where those numbers are coming from.

Well, I guess I haven't been able to follow your math on this. I had thought that, aside from variations due to rounding to whole numbers or anomalies arising from a situation in which there is a very small number of voters, the fraction of voters who can elect at least one candidate of their choice by coordinating their cumulative votes remains the same regardless of how many voters there are in total.

I don't think the answer is so clear-cut as that. Suppose the bylaws said that the order of business shall be as prescribed in Robert's Rules of Order Newly Revised. Would you say that the society could still adopt a special order of business to govern all its meetings because RONR says that it can?

*Or maybe it was a four-fifths vote discussed in PL?

I think I am lost at this point, so let's go back to an earlier question. You had asked: "So what would you say is the vote required to suspend a rule requiring a three-fourths vote for the adoption of a particular motion?" My answer is a three-fourths vote. I believe this question is discussed in PL with regard to a rule requiring a three-fourths vote* to allow non-members on the floor. My copy of PL is currently further than my elbow (not counting any online edition that I could access with my thumbs), but my recollection is that the General says this rule could only be suspended by a three-fourths vote (so that there would be little reason to adopt such a suspension of the rules). Now suppose that instead of a motion to suspend the rules and admit a non-member to the floor, a motion (for sake of discussion, let's call it a main motion) to admit a non-member is already pending, and someone moves to suspend the rules in order that the pending motion can be adopted by a two-thirds vote. I would say that this motion also requires a three-fourths vote. However, I must be missing something about what you've been saying (about fifty times), because if we need to know in advance how many members will vote on the main motion, then there is not actually any particular size minority protected by the rule?

Yes, they gave up their rights to have the election conducted by cumulative voting.

You can't abstain on a motion to give up your rights and then claim the same rights after the motion is legitimately adopted.

We know that no more members can vote in the election than the number who are present at the time of the election. And for purposes of suspending the rules relating to the number of votes necessary for election, I think this number (the number present) is adequate to work with, because as I pointed out to J. J. earlier, there's nothing preventing the assembly from actually conducting the election at that time, so any consideration with regard to the rights of absent members is not relevant. And I further think that the number present and voting is adequate to work with, because any members who abstain thereby relinquish any rights given up by adoption of the motion to suspend the rules.

But we know what the minimum is to guarantee it. And presumably the members who are present and are eligible to vote in the election know how they intend to vote in the election, and therefore know what to do to protect their rights in voting in the election. The motion to suspend the rules relating to the election becomes part of the election process, and that motion cannot violate their rights either.

But if the fraction of members voting in opposition on the motion to suspend the rules is as large as the fraction protected by the rule governing the election, then it seems obvious to me this should be sufficient to prevent the suspension of the rules. Conversely, if sufficient members eligible to vote in the election either abstain or vote in favor of suspending the rules, then this may be taken as a relinquishment of their rights to achieve their desired outcome in the election by the particular fraction in question.

Robert's Rules does not give members any such right. However, it is common for statutes to require the opposition of a member of the board of directors to be noted in the minutes in certain situations, so that might be what this member was thinking of.

And don't forget that it requires a two-thirds vote on the motion for the Previous Question in order to close the debate. If this motion does not receive a two-thirds vote, then debate continues just as if the motion had not been made.

Are you sure that you have quoted exactly what it says? Without some further context it's difficult to understand what this is intended to mean.

Well then I have to disagree. I don't see what the point would be in having a rule that enfranchises the members in a certain way, if 2/3 of the members voting could get their desired outcome by suspending the rules when they could not achieve it directly.

Well I still think there is something fishy about suspending this rule. If this method of electing the committee were prescribed directly in the bylaws, would you still say that the rules could be suspended so as to disallow cumulative voting?

That's an interesting point, but I don't see why it should make any logical difference.

I do not think that the rule in 46:43 protects "a minority of a particular size" as contemplated by the rule in 25:2(7). In that case, maybe the rule cannot be suspended at all, as suspending the rule would take away a member's right to vote (as many times as he is entitled to).