Why ballot box commentary cannot work

This is a follow-up to my report yesterday that the Green Party of Canada has added partisan messaging to their ballots for this year’s convention. It’s not necessary to read that post first, but it provides a real-world example of this phenomenon, if you would like to see one.

For anybody who has even briefly observed the differences between democratic and authoritarian states, it should be obvious that democracies try to keep their voting processes strictly non-partisan, while authoritarian states typically do the opposite.

But just in case it’s not obvious why ballot box commentary is a terrible idea for any organization, this post explains the problems.

Short summaries are inherently biased

Words contain information, but not unlimited amounts of information. When commenting on a complicated idea, such as a policy proposal, technical standard, and so on, the important features of that idea simply cannot be condensed into a two-sentence comment. Trying to do so will result in information being lost.

As a result, even a completely unbiased person cannot write an unbiased comment if the length limit is less than the complexity of the idea being discussed. They will be forced to discard information.

I happen to have a lot of first-hand experience with this phenomenon, because part of my work involves writing book synopses of different lengths for marketing purposes. Typically, we start with the longest synopsis, around 200 words. Then we write shorter and shorter versions, the shortest being around 20 words, editing out detail as we go.

There is an unavoidable point in this process where some key element of the book—a central character, plot event, or whatever—is discarded, and the synopsis becomes something less than the true whole of the book. For marketing purposes, this is okay, as long as the short synopsis still hooks people. As part of a voting process, this loss of information is not okay.

Limited space is a scarce resource; this creates conflict

Scarce resources create conflicts between people who want the resource used in different ways. In the case of ballot box comments, this has several negative effects:

• The people writing or curating the ballot comments are placed in a conflict of interest (assuming they are members), since they have a stake in the direction and success of the organization.
• Members whose proposals receive negative comments may resent and mistrust the system. Some will challenge the validity of the comments; the continued presence of these challenges will reduce trust among all members.
• Members whose proposals receive positive comments may be resented by their peers or seen as having received special favours. This creates suspicion between members.
• When a proposal receives a positive comment and passes by a slim margin, or receives a negative comment and fails by a slim margin, especially if the proposal is contentious, members may believe the comment corrupted the result.

All of these problems stem from unnecessarily creating a scarce resource (ballot box comment space) which does not need to exist. Removing ballot box comments immediately makes this conflict and negativity impossible.

The authority-trust paradox

If a ballot is going to include comments, someone must write or curate those comments. To maintain members’ faith in the system, the organization will almost certainly select a person or group that is seen as trustworthy.

This creates a paradox. The more respected and trusted a person you pick to write the comments, the more members will trust their judgment implicitly, and the more those comments will distort the voting result. The more the result is distorted, the less trust members will ultimately have in the system, as they witness results that are swayed by the ballot-box comments.

There is a second paradox. The more respected and trusted a person you pick, the more likely they are to already be in a position of authority within the party. The more authority they already possess, the more authoritarian a system you create by giving them power over the voting process.

Last-minute failures cannot be corrected

Any well-designed system assumes that both errors and abuse will occur, and includes a process to detect and correct the errors and abuse.

With ballot-box commentary this is impossible, because any errors or abuse will be detected during the voting process, meaning that any reports of errors or abuse will be received after some or all votes have already been received.

(Note: This problem cannot be solved by displaying the ballot-box comments ahead of time, because an error or abuse of process could still occur when copying those comments over to the final ballot.)

An organization that employs ballot-box commentary will eventually face a situation where it must either overturn a vote, or accept a tainted result. These situations are extremely dangerous to the internal fabric of the organization, because they always leave significant numbers of people very upset:

• If a new vote is not called, anybody who saw the initial vote as tainted will be upset.
• If a new vote is called, and the result changes, anybody who saw the initial vote as legitimate will be upset.
• If a new vote is called, and the result does not change, anybody who saw the initial vote as tainted will be suspicious that the second vote was distorted by reaction to the first vote.

There are other permutations, but the point here is, somebody always goes home extremely unhappy. A well-run organization never risks creating these situations.

Last-minute summaries suppress democratic engagement

If a members knows that the ballot will include comments that summarize the pros and cons of a proposal, they are less likely to engage in a full debate process. This is bad for the health of the organization.

Some people would argue that it’s better to have those members at least be somewhat informed through ballot box comments, otherwise they will vote for whatever superficially sounds good.

This will sound counter-intuitive, but superficial voting is still more democratic than directed voting. The ways in which a proposal can attract superficial votes—for example, by having an appealing name or a really catchy summary—are achieved through skills that anybody can practice, ask for help with, or lend to others. Proposals that have the most widespread support are more likely to attract this help prior to the vote. This makes “voting for what sounds good” inherently more democratic than “vote for whatever the ballot box comment says”.

Risk of authoritarian takeover

Finally, there is a serious risk in introducing commentary into ballots, which is that it creates another tool for authoritarian leaders to take over an organization.

When a voting system is neutral and there is an expectation that it remain neutral, any attempt to introduce non-neutral elements will be easily seen and provoke a reaction (hopefully a strong one).

When a voting system already contains non-neutral elements, it is much easier to introduce more non-neutral elements, and to make them more and more partisan, until the system is corrupted to support an authoritarian leader.

Any organization that has value (money, reputation, members, etc) must vigilantly guard against people who would take over the organization and use it for their own personal ends. Keeping internal processes non-partisan is a necessary part of this vigilance.

Summary

In any organization, flawed processes are unavoidable, but it’s foolish to add additional flaws with no good reason.

Ballot-box commentary is a solution in search of a problem that doesn’t exist. The risks are significant: internal conflict, loss of trust, suppressed engagement, potential “disaster scenarios” due to errors or abuse, and the potential for enabling authoritarian takeover. The goal—giving voters key information on proposals—can be achieved other ways with far less risk and more positive side benefits.

Organizations that want to be democratic should avoid ballot-box commentary and keep the voting process strictly non-partisan. Instead, to encourage more informed voting, the organization should develop, facilitate, and continually nurture free and open internal debates as part of its organizational culture.